"Bod" wrote in message
...
On 17/10/2016 18:24, Rod Speed wrote:


"Bod" wrote in message
...
On 17/10/2016 17:29, Clive George wrote:
On 17/10/2016 16:14, Bod wrote:
On 17/10/2016 16:06, Clive George wrote:
On 17/10/2016 15:12, Bod wrote:
"Afford the luxury"??

If your system is running into swap, it will be slower than it
needs to
be. It's not so much affording the luxury, as affording the basics.

TNP might be happy running gridwatch on a cheese-pared setup, but
memory
is really very cheap these days, so anybody doing this sort of
thing
for
a living will do it properly - which for most servers means
running in
memory. (there are probably exceptions, I've just not hit them yet,
but
I do have a fairly wide experience of servers...)

The price depends on what type of memory. 32GB of DDR4 2600 mhz
for my
laptop is £171. I wouldn't call that cheap.

If you're doing this for a living and running something which needs
32GB
of memory, 171 quid is cheap. You're neither doing it for a living
nor
do you need anything like that, so you think it's expensive.

You probably also think over 500 quid for a laptop is expensive :-)

(basic reasoning is pretty much that if you're doing it for a living,
other costs will dwarf it, and if you need it you'll either be making
rather more than that or saving enough time to make it worth it.)

Has TNP posted how much his gridwatch server has? For comparison, I
think the cheapest lowest spec Azure VM is 3.5GB, and that will all
be
available as full speed memory if you use it.


My latest laptop cost £900.

And you're claiming 171 quid for a significant upgrade isn't cheap?

It's subjective. I could've got a much cheaper i7 laptop for about £550.

So it was stupid to have just 8GB of ram and to close apps when you stop
using them.

And you ****ed more against the wall on the SSD than you would
have spent on the right amount of physical ram in that laptop too.

Lol.

Village eejut imitations cut no mustard around here, child.

On Mon, 17 Oct 2016 14:57:55 +0100, Clive George
wrote:

snip

TNP might be happy running gridwatch on a cheese-pared setup, but memory
is really very cheap these days, so anybody doing this sort of thing for
a living will do it properly - which for most servers means running in
memory. (there are probably exceptions, I've just not hit them yet, but
I do have a fairly wide experience of servers...)

I remember going though the calculations with the delegates for RAM
for a Netware 3.12 Server. So much for each user, more for each .nlm
and each namespace supported etc and then once you come up with the
figure, probably fit 10x that because you could. ;-)

Cheers, T i m

On 17/10/2016 18:30, Rod Speed wrote:


"Bod" wrote in message
...
On 17/10/2016 18:00, Rod Speed wrote:


"Bod" wrote in message
...
On 17/10/2016 13:37, Clive George wrote:
On 15/10/2016 15:03, David wrote:
On Sat, 15 Oct 2016 10:51:41 +0100, The Natural Philosopher wrote:

On 15/10/16 10:33, Adrian Caspersz wrote:
snip

Faster booting, faster program load and, if memory is short faster
access to swap if that's on the SSD (although if you want to wear
out an
SSD, using it for swap is one of the best ways).
snip

Just pondering on SSDs and Swap.

If you have a system with just an SSD then what do you use as swap?

Assuming that you don't (as in most laptops and older desktops) have
the
space for masses of memory?

"Most laptops"? 8G and 16G on the laptops in use here :-)

For some years now I've made sure no system I have anything to do with
is running into swap.

I disable the swap and never run out of ram for what I do and I only
use 8GB of ram.

And you stupidly close apps when you stop using them for a bit,
and that is why you see a better result with an SSD. If you werent
that stupid, you wouldnt see any better result with an SSD if you have
enough of a clue to not turn the system off when you arent using it.

It's left on all day. Only gets switched off at bedtime,

And doing that is stupid.

Why would I leave it turned on all night when it's not being used.

Because it starts instantly next day, stupid.

Lol. Don't talk daft. I get up, press button to start computer, make a
cup of tea, get online, finish making tea, then onto the computer.

Tell me how much time I am waiting more than you?

On 17/10/2016 18:32, Rod Speed wrote:


"Bod" wrote in message
...
On 17/10/2016 18:16, Rod Speed wrote:


"Bod" wrote in message
...
On 17/10/2016 16:06, Clive George wrote:
On 17/10/2016 15:12, Bod wrote:
"Afford the luxury"??

If your system is running into swap, it will be slower than it
needs to
be. It's not so much affording the luxury, as affording the basics.

TNP might be happy running gridwatch on a cheese-pared setup, but
memory
is really very cheap these days, so anybody doing this sort of
thing for
a living will do it properly - which for most servers means
running in
memory. (there are probably exceptions, I've just not hit them yet,
but
I do have a fairly wide experience of servers...)

The price depends on what type of memory. 32GB of DDR4 2600 mhz
for my
laptop is £171. I wouldn't call that cheap.

If you're doing this for a living and running something which needs
32GB
of memory, 171 quid is cheap. You're neither doing it for a living nor
do you need anything like that, so you think it's expensive.

You probably also think over 500 quid for a laptop is expensive :-)

(basic reasoning is pretty much that if you're doing it for a living,
other costs will dwarf it, and if you need it you'll either be making
rather more than that or saving enough time to make it worth it.)

Has TNP posted how much his gridwatch server has? For comparison, I
think the cheapest lowest spec Azure VM is 3.5GB, and that will all be
available as full speed memory if you use it.


My latest laptop cost £900.

So it was stupid to not have more than 8GB of ram
in it and to close apps when you stop using them.

Because that's the amount that was in it when I bought it.
I'm gonna get another 24GB of ram anyway.

So you ****ed your money against the wall on the SSD
and will get no advantage from it once you have 32GB
of ram, dont close apps when you stop using them and
dont stupidly turn the system off every day.

Tell me how fast you can load a pagefull of, say, photos from a
mechanical drive?

On 17/10/2016 19:16, Bod wrote:
On 17/10/2016 18:32, Rod Speed wrote:


"Bod" wrote in message
...
On 17/10/2016 18:16, Rod Speed wrote:


"Bod" wrote in message
...
On 17/10/2016 16:06, Clive George wrote:
On 17/10/2016 15:12, Bod wrote:
"Afford the luxury"??

If your system is running into swap, it will be slower than it
needs to
be. It's not so much affording the luxury, as affording the basics.

TNP might be happy running gridwatch on a cheese-pared setup, but
memory
is really very cheap these days, so anybody doing this sort of
thing for
a living will do it properly - which for most servers means
running in
memory. (there are probably exceptions, I've just not hit them yet,
but
I do have a fairly wide experience of servers...)

The price depends on what type of memory. 32GB of DDR4 2600 mhz
for my
laptop is £171. I wouldn't call that cheap.

If you're doing this for a living and running something which needs
32GB
of memory, 171 quid is cheap. You're neither doing it for a living
nor
do you need anything like that, so you think it's expensive.

You probably also think over 500 quid for a laptop is expensive :-)

(basic reasoning is pretty much that if you're doing it for a living,
other costs will dwarf it, and if you need it you'll either be making
rather more than that or saving enough time to make it worth it.)

Has TNP posted how much his gridwatch server has? For comparison, I
think the cheapest lowest spec Azure VM is 3.5GB, and that will
all be
available as full speed memory if you use it.


My latest laptop cost £900.

So it was stupid to not have more than 8GB of ram
in it and to close apps when you stop using them.

Because that's the amount that was in it when I bought it.
I'm gonna get another 24GB of ram anyway.

So you ****ed your money against the wall on the SSD
and will get no advantage from it once you have 32GB
of ram, dont close apps when you stop using them and
dont stupidly turn the system off every day.

Tell me how fast you can load a pagefull of, say, photos from a
mechanical drive?
*displaying large icons*
I click on my SSD and they all show within a fraction of a second.

"Bod" wrote in message
...
On 17/10/2016 18:30, Rod Speed wrote:


"Bod" wrote in message
...
On 17/10/2016 18:00, Rod Speed wrote:


"Bod" wrote in message
...
On 17/10/2016 13:37, Clive George wrote:
On 15/10/2016 15:03, David wrote:
On Sat, 15 Oct 2016 10:51:41 +0100, The Natural Philosopher wrote:

On 15/10/16 10:33, Adrian Caspersz wrote:
snip

Faster booting, faster program load and, if memory is short faster
access to swap if that's on the SSD (although if you want to wear
out an
SSD, using it for swap is one of the best ways).
snip

Just pondering on SSDs and Swap.

If you have a system with just an SSD then what do you use as swap?

Assuming that you don't (as in most laptops and older desktops) have
the
space for masses of memory?

"Most laptops"? 8G and 16G on the laptops in use here :-)

For some years now I've made sure no system I have anything to do
with
is running into swap.

I disable the swap and never run out of ram for what I do and I only
use 8GB of ram.

And you stupidly close apps when you stop using them for a bit,
and that is why you see a better result with an SSD. If you werent
that stupid, you wouldnt see any better result with an SSD if you have
enough of a clue to not turn the system off when you arent using it.

It's left on all day. Only gets switched off at bedtime,

And doing that is stupid.

Why would I leave it turned on all night when it's not being used.

Because it starts instantly next day, stupid.

Lol. Don't talk daft.

Nothing daft about it.

I get up, press button to start computer, make a cup of tea, get online,
finish making tea, then onto the computer.

And if you werent actually stupid enough to do it like that,
you could just tap the touchpad and have it work instantly.

"Bod" wrote in message
...
On 17/10/2016 18:32, Rod Speed wrote:


"Bod" wrote in message
...
On 17/10/2016 18:16, Rod Speed wrote:


"Bod" wrote in message
...
On 17/10/2016 16:06, Clive George wrote:
On 17/10/2016 15:12, Bod wrote:
"Afford the luxury"??

If your system is running into swap, it will be slower than it
needs to
be. It's not so much affording the luxury, as affording the basics.

TNP might be happy running gridwatch on a cheese-pared setup, but
memory
is really very cheap these days, so anybody doing this sort of
thing for
a living will do it properly - which for most servers means
running in
memory. (there are probably exceptions, I've just not hit them yet,
but
I do have a fairly wide experience of servers...)

The price depends on what type of memory. 32GB of DDR4 2600 mhz
for my
laptop is £171. I wouldn't call that cheap.

If you're doing this for a living and running something which needs
32GB
of memory, 171 quid is cheap. You're neither doing it for a living
nor
do you need anything like that, so you think it's expensive.

You probably also think over 500 quid for a laptop is expensive :-)

(basic reasoning is pretty much that if you're doing it for a living,
other costs will dwarf it, and if you need it you'll either be making
rather more than that or saving enough time to make it worth it.)

Has TNP posted how much his gridwatch server has? For comparison, I
think the cheapest lowest spec Azure VM is 3.5GB, and that will all
be
available as full speed memory if you use it.


My latest laptop cost £900.

So it was stupid to not have more than 8GB of ram
in it and to close apps when you stop using them.

Because that's the amount that was in it when I bought it.
I'm gonna get another 24GB of ram anyway.

So you ****ed your money against the wall on the SSD
and will get no advantage from it once you have 32GB
of ram, dont close apps when you stop using them and
dont stupidly turn the system off every day.

Tell me how fast you can load a pagefull of, say, photos from a mechanical
drive?

Fast enough to not bother to **** any money against
the wall on an SSD. I do that on the smartphone I took
the photos with anyway, not stupid enough to do that
on the desktop or laptop and its instant on the smartphone.

On 17/10/16 14:12, Bod wrote:
On 17/10/2016 13:46, The Natural Philosopher wrote:
On 17/10/16 13:37, Clive George wrote:
On 15/10/2016 15:03, David wrote:
On Sat, 15 Oct 2016 10:51:41 +0100, The Natural Philosopher wrote:

On 15/10/16 10:33, Adrian Caspersz wrote:
snip

Faster booting, faster program load and, if memory is short faster
access to swap if that's on the SSD (although if you want to wear
out an
SSD, using it for swap is one of the best ways).
snip

Just pondering on SSDs and Swap.

If you have a system with just an SSD then what do you use as swap?

Assuming that you don't (as in most laptops and older desktops) have
the
space for masses of memory?

"Most laptops"? 8G and 16G on the laptops in use here :-)

For some years now I've made sure no system I have anything to do with
is running into swap.


Nice if you can afford the luxury....

Guess how much RAM gridwatch runs on, and how much swap it has?


How much?

384Mbyte

Think there's a gig of swap.

--
"The great thing about Glasgow is that if there's a nuclear attack it'll
look exactly the same afterwards."

Billy Connolly

On 17/10/16 14:57, Clive George wrote:
On 17/10/2016 14:12, Bod wrote:
On 17/10/2016 13:46, The Natural Philosopher wrote:
On 17/10/16 13:37, Clive George wrote:
On 15/10/2016 15:03, David wrote:
On Sat, 15 Oct 2016 10:51:41 +0100, The Natural Philosopher wrote:

On 15/10/16 10:33, Adrian Caspersz wrote:
snip

Faster booting, faster program load and, if memory is short faster
access to swap if that's on the SSD (although if you want to wear
out an
SSD, using it for swap is one of the best ways).
snip

Just pondering on SSDs and Swap.

If you have a system with just an SSD then what do you use as swap?

Assuming that you don't (as in most laptops and older desktops) have
the
space for masses of memory?

"Most laptops"? 8G and 16G on the laptops in use here :-)

For some years now I've made sure no system I have anything to do with
is running into swap.


Nice if you can afford the luxury....

Guess how much RAM gridwatch runs on, and how much swap it has?


How much?

"Afford the luxury"??

If your system is running into swap, it will be slower than it needs to
be. It's not so much affording the luxury, as affording the basics.

TNP might be happy running gridwatch on a cheese-pared setup, but memory
is really very cheap these days, so anybody doing this sort of thing for
a living will do it properly - which for most servers means running in
memory. (there are probably exceptions, I've just not hit them yet, but
I do have a fairly wide experience of servers...)

Memory in a shared host virtual machine is NOT cheap :-(


--
"Women actually are capable of being far more than the feminists will
let them."

On 17/10/16 17:51, Mike Tomlinson wrote:
En el artÃ*culo ,
Clive George escribió:

TNP might be happy running gridwatch on a cheese-pared setup

He's running it on a cheapo hosted setup via Paragon Internet.

No, I am not


--
"Women actually are capable of being far more than the feminists will
let them."

"Bod" wrote in message
...
On 17/10/2016 19:16, Bod wrote:
On 17/10/2016 18:32, Rod Speed wrote:


"Bod" wrote in message
...
On 17/10/2016 18:16, Rod Speed wrote:


"Bod" wrote in message
...
On 17/10/2016 16:06, Clive George wrote:
On 17/10/2016 15:12, Bod wrote:
"Afford the luxury"??

If your system is running into swap, it will be slower than it
needs to
be. It's not so much affording the luxury, as affording the
basics.

TNP might be happy running gridwatch on a cheese-pared setup, but
memory
is really very cheap these days, so anybody doing this sort of
thing for
a living will do it properly - which for most servers means
running in
memory. (there are probably exceptions, I've just not hit them
yet,
but
I do have a fairly wide experience of servers...)

The price depends on what type of memory. 32GB of DDR4 2600 mhz
for my
laptop is £171. I wouldn't call that cheap.

If you're doing this for a living and running something which needs
32GB
of memory, 171 quid is cheap. You're neither doing it for a living
nor
do you need anything like that, so you think it's expensive.

You probably also think over 500 quid for a laptop is expensive :-)

(basic reasoning is pretty much that if you're doing it for a
living,
other costs will dwarf it, and if you need it you'll either be
making
rather more than that or saving enough time to make it worth it.)

Has TNP posted how much his gridwatch server has? For comparison, I
think the cheapest lowest spec Azure VM is 3.5GB, and that will
all be
available as full speed memory if you use it.


My latest laptop cost £900.

So it was stupid to not have more than 8GB of ram
in it and to close apps when you stop using them.

Because that's the amount that was in it when I bought it.
I'm gonna get another 24GB of ram anyway.

So you ****ed your money against the wall on the SSD
and will get no advantage from it once you have 32GB
of ram, dont close apps when you stop using them and
dont stupidly turn the system off every day.

Tell me how fast you can load a pagefull of, say, photos from a
mechanical drive?
*displaying large icons*
I click on my SSD and they all show within a fraction of a second.

I'm not stupid enough to look at them that way on the desktop/laptop.

I do that on the smartphone I took them on and they are
instantly available there and are much more convenient
to send to anywhere I need to send them from there.

On 17/10/2016 19:38, Rod Speed wrote:


"Bod" wrote in message
...
On 17/10/2016 18:30, Rod Speed wrote:


"Bod" wrote in message
...
On 17/10/2016 18:00, Rod Speed wrote:


"Bod" wrote in message
...
On 17/10/2016 13:37, Clive George wrote:
On 15/10/2016 15:03, David wrote:
On Sat, 15 Oct 2016 10:51:41 +0100, The Natural Philosopher wrote:

On 15/10/16 10:33, Adrian Caspersz wrote:
snip

Faster booting, faster program load and, if memory is short faster
access to swap if that's on the SSD (although if you want to wear
out an
SSD, using it for swap is one of the best ways).
snip

Just pondering on SSDs and Swap.

If you have a system with just an SSD then what do you use as swap?

Assuming that you don't (as in most laptops and older desktops)
have
the
space for masses of memory?

"Most laptops"? 8G and 16G on the laptops in use here :-)

For some years now I've made sure no system I have anything to do
with
is running into swap.

I disable the swap and never run out of ram for what I do and I only
use 8GB of ram.

And you stupidly close apps when you stop using them for a bit,
and that is why you see a better result with an SSD. If you werent
that stupid, you wouldnt see any better result with an SSD if you
have
enough of a clue to not turn the system off when you arent using it.

It's left on all day. Only gets switched off at bedtime,

And doing that is stupid.

Why would I leave it turned on all night when it's not being used.

Because it starts instantly next day, stupid.

Lol. Don't talk daft.

Nothing daft about it.

I get up, press button to start computer, make a cup of tea, get
online, finish making tea, then onto the computer.

And if you werent actually stupid enough to do it like that,
you could just tap the touchpad and have it work instantly.


That is being stupid!!?
Get a life, Rod.

On 17/10/2016 19:41, Rod Speed wrote:


"Bod" wrote in message
...
On 17/10/2016 18:32, Rod Speed wrote:


"Bod" wrote in message
...
On 17/10/2016 18:16, Rod Speed wrote:


"Bod" wrote in message
...
On 17/10/2016 16:06, Clive George wrote:
On 17/10/2016 15:12, Bod wrote:
"Afford the luxury"??

If your system is running into swap, it will be slower than it
needs to
be. It's not so much affording the luxury, as affording the
basics.

TNP might be happy running gridwatch on a cheese-pared setup, but
memory
is really very cheap these days, so anybody doing this sort of
thing for
a living will do it properly - which for most servers means
running in
memory. (there are probably exceptions, I've just not hit them
yet,
but
I do have a fairly wide experience of servers...)

The price depends on what type of memory. 32GB of DDR4 2600 mhz
for my
laptop is £171. I wouldn't call that cheap.

If you're doing this for a living and running something which needs
32GB
of memory, 171 quid is cheap. You're neither doing it for a
living nor
do you need anything like that, so you think it's expensive.

You probably also think over 500 quid for a laptop is expensive :-)

(basic reasoning is pretty much that if you're doing it for a
living,
other costs will dwarf it, and if you need it you'll either be
making
rather more than that or saving enough time to make it worth it.)

Has TNP posted how much his gridwatch server has? For comparison, I
think the cheapest lowest spec Azure VM is 3.5GB, and that will
all be
available as full speed memory if you use it.


My latest laptop cost £900.

So it was stupid to not have more than 8GB of ram
in it and to close apps when you stop using them.

Because that's the amount that was in it when I bought it.
I'm gonna get another 24GB of ram anyway.

So you ****ed your money against the wall on the SSD
and will get no advantage from it once you have 32GB
of ram, dont close apps when you stop using them and
dont stupidly turn the system off every day.

Tell me how fast you can load a pagefull of, say, photos from a
mechanical drive?

Fast enough to not bother to **** any money against
the wall on an SSD. I do that on the smartphone I took
the photos with anyway, not stupid enough to do that
on the desktop or laptop and its instant on the smartphone.


"Hard drives win on price, capacity, and availability. SSDs work best if
speed, ruggedness, form factor, noise, or fragmentation (technically
part of speed) are important factors to you. If it weren't for the price
and capacity issues, SSDs would be the winner hands down".

http://uk.pcmag.com/storage-devices-...the-difference

On 17/10/2016 19:46, The Natural Philosopher wrote:
On 17/10/16 14:12, Bod wrote:
On 17/10/2016 13:46, The Natural Philosopher wrote:
On 17/10/16 13:37, Clive George wrote:
On 15/10/2016 15:03, David wrote:
On Sat, 15 Oct 2016 10:51:41 +0100, The Natural Philosopher wrote:

On 15/10/16 10:33, Adrian Caspersz wrote:
snip

Faster booting, faster program load and, if memory is short faster
access to swap if that's on the SSD (although if you want to wear
out an
SSD, using it for swap is one of the best ways).
snip

Just pondering on SSDs and Swap.

If you have a system with just an SSD then what do you use as swap?

Assuming that you don't (as in most laptops and older desktops) have
the
space for masses of memory?

"Most laptops"? 8G and 16G on the laptops in use here :-)

For some years now I've made sure no system I have anything to do with
is running into swap.


Nice if you can afford the luxury....

Guess how much RAM gridwatch runs on, and how much swap it has?


How much?

384Mbyte

Think there's a gig of swap.

People often think they need more ram than they actually do.

On 17/10/2016 12:34, The Natural Philosopher wrote:
On 17/10/16 11:31, John Rumm wrote:

It makes far more sense to ensure the partitions are aligned so that the
OS allocation unit is on a 4K boundary (which is the default action on a
modern OS anyway)

My point is that that is what the SSD firmware will in fact do.

No matter where the partition boundary is.

Could you provide a link to a manufacturers documentation for this since
I have not yet seen any SSDs that claim to be able to remap misaligned
partitions?


--
Cheers,

John.

/================================================== ===============\
| Internode Ltd - http://www.internode.co.uk |
|-----------------------------------------------------------------|
| John Rumm - john(at)internode(dot)co(dot)uk |
\================================================= ================/

On 17/10/2016 11:42, T i m wrote:
On Mon, 17 Oct 2016 11:31:48 +0100, John Rumm
wrote:

On 17/10/2016 10:45, The Natural Philosopher wrote:
On 17/10/16 09:37, John Rumm wrote:
In reality it could be anywhere on any boundary the
SSD firmware and processor puts it.

Not quite - there is not a one to one mapping of OS allocation units to
flash pages. For optimum performance you need to ensure that whichever
allocation unit you update, the SSD can do that update by operating on
one (and only one) page of flash. With the wrong alignment, you can end
up with the SSD needing to do two page updates for each OS allocation
unit update.


However all that may be true, but its not under user level control via
partioning, and its handled internally by the SSD.

There is no obvious way a SSD could make a sensible choice to internally
remap alignment if it turns out you have managed to install an OS
partition with a start LBA offset from the ideal. Especially as one
physical drive can host more than one partition, and if you really
tried, you could end up with several partitions each with different
alignments.

It makes far more sense to ensure the partitions are aligned so that the
OS allocation unit is on a 4K boundary (which is the default action on a
modern OS anyway)

So and irrespective of any performance impact ITRW, if some software
(Gparted) can see and display that these alignments aren't made *and*
can set them, is Gparted actually then *actually / physically*
resetting said alignments or just indicating it is?

It would need to copy the entire partition up or down a number of
sectors. (However the OS may bork and need a reinstall if you do!)

What would be a good (valid) way of checking for such things
(increased performance hopefully) pre and post adjustment?

On windows, do a msinfo32 in a command prompt to start the system
information tool. Then expand out the Components - Storage - Disks
part of the tree view. That will show all the physical drives, and the
partitions and their starting offsets. Divide the offset by 4096 and
hope you get an integer answer ;-)

e.g:

Description Disk drive
Manufacturer (Standard disk drives)
Model KINGSTON SHSS37A240G ATA Device
Bytes/Sector 512
Media Loaded Yes
Media Type Fixed hard disk
Partitions 1
SCSI Bus 0
SCSI Logical Unit 0
SCSI Port 0
SCSI Target ID 0
Sectors/Track 63
Size 223.57 GB (240,054,796,800 bytes)
Total Cylinders 29,185
Total Sectors 468,857,025
Total Tracks 7,442,175
Tracks/Cylinder 255
Partition Disk #0, Partition #0
Partition Size 223.57 GB (240,055,746,560 bytes)
Partition Starting Offset 1,048,576 bytes

It all sounds like re-numbering sector to reduce latency with Optune
all those years ago. ;-)

Yup, when disc controllers were so slow they might not be able to read
consecutive sectors if they were physically adjacent to each other!


--
Cheers,

John.

/================================================== ===============\
| Internode Ltd - http://www.internode.co.uk |
|-----------------------------------------------------------------|
| John Rumm - john(at)internode(dot)co(dot)uk |
\================================================= ================/

"Bod" wrote in message
...
On 17/10/2016 19:38, Rod Speed wrote:


"Bod" wrote in message
...
On 17/10/2016 18:30, Rod Speed wrote:


"Bod" wrote in message
...
On 17/10/2016 18:00, Rod Speed wrote:


"Bod" wrote in message
...
On 17/10/2016 13:37, Clive George wrote:
On 15/10/2016 15:03, David wrote:
On Sat, 15 Oct 2016 10:51:41 +0100, The Natural Philosopher wrote:

On 15/10/16 10:33, Adrian Caspersz wrote:
snip

Faster booting, faster program load and, if memory is short
faster
access to swap if that's on the SSD (although if you want to wear
out an
SSD, using it for swap is one of the best ways).
snip

Just pondering on SSDs and Swap.

If you have a system with just an SSD then what do you use as
swap?

Assuming that you don't (as in most laptops and older desktops)
have
the
space for masses of memory?

"Most laptops"? 8G and 16G on the laptops in use here :-)

For some years now I've made sure no system I have anything to do
with
is running into swap.

I disable the swap and never run out of ram for what I do and I only
use 8GB of ram.

And you stupidly close apps when you stop using them for a bit,
and that is why you see a better result with an SSD. If you werent
that stupid, you wouldnt see any better result with an SSD if you
have
enough of a clue to not turn the system off when you arent using it.

It's left on all day. Only gets switched off at bedtime,

And doing that is stupid.

Why would I leave it turned on all night when it's not being used.

Because it starts instantly next day, stupid.

Lol. Don't talk daft.

Nothing daft about it.

I get up, press button to start computer, make a cup of tea, get
online, finish making tea, then onto the computer.

And if you werent actually stupid enough to do it like that,
you could just tap the touchpad and have it work instantly.


That is being stupid!!?

Is that a question or and exclamation, stupid ?

Get a life, Rod.

Get a viable response, gutless.

"Bod" wrote in message
...
On 17/10/2016 19:41, Rod Speed wrote:


"Bod" wrote in message
...
On 17/10/2016 18:32, Rod Speed wrote:


"Bod" wrote in message
...
On 17/10/2016 18:16, Rod Speed wrote:


"Bod" wrote in message
...
On 17/10/2016 16:06, Clive George wrote:
On 17/10/2016 15:12, Bod wrote:
"Afford the luxury"??

If your system is running into swap, it will be slower than it
needs to
be. It's not so much affording the luxury, as affording the
basics.

TNP might be happy running gridwatch on a cheese-pared setup, but
memory
is really very cheap these days, so anybody doing this sort of
thing for
a living will do it properly - which for most servers means
running in
memory. (there are probably exceptions, I've just not hit them
yet,
but
I do have a fairly wide experience of servers...)

The price depends on what type of memory. 32GB of DDR4 2600 mhz
for my
laptop is £171. I wouldn't call that cheap.

If you're doing this for a living and running something which needs
32GB
of memory, 171 quid is cheap. You're neither doing it for a
living nor
do you need anything like that, so you think it's expensive.

You probably also think over 500 quid for a laptop is expensive :-)

(basic reasoning is pretty much that if you're doing it for a
living,
other costs will dwarf it, and if you need it you'll either be
making
rather more than that or saving enough time to make it worth it.)

Has TNP posted how much his gridwatch server has? For comparison, I
think the cheapest lowest spec Azure VM is 3.5GB, and that will
all be
available as full speed memory if you use it.


My latest laptop cost £900.

So it was stupid to not have more than 8GB of ram
in it and to close apps when you stop using them.

Because that's the amount that was in it when I bought it.
I'm gonna get another 24GB of ram anyway.

So you ****ed your money against the wall on the SSD
and will get no advantage from it once you have 32GB
of ram, dont close apps when you stop using them and
dont stupidly turn the system off every day.

Tell me how fast you can load a pagefull of, say, photos from a
mechanical drive?

Fast enough to not bother to **** any money against
the wall on an SSD. I do that on the smartphone I took
the photos with anyway, not stupid enough to do that
on the desktop or laptop and its instant on the smartphone.


"Hard drives win on price, capacity, and availability. SSDs work best if
speed, ruggedness, form factor, noise, or fragmentation (technically part
of speed) are important factors to you. If it weren't for the price and
capacity issues, SSDs would be the winner hands down".

http://uk.pcmag.com/storage-devices-...the-difference

Irrelevant to what is being discussed.

On 17/10/2016 20:07, Bod wrote:
On 17/10/2016 19:46, The Natural Philosopher wrote:
On 17/10/16 14:12, Bod wrote:
On 17/10/2016 13:46, The Natural Philosopher wrote:
On 17/10/16 13:37, Clive George wrote:
On 15/10/2016 15:03, David wrote:
On Sat, 15 Oct 2016 10:51:41 +0100, The Natural Philosopher wrote:

On 15/10/16 10:33, Adrian Caspersz wrote:
snip

Faster booting, faster program load and, if memory is short faster
access to swap if that's on the SSD (although if you want to wear
out an
SSD, using it for swap is one of the best ways).
snip

Just pondering on SSDs and Swap.

If you have a system with just an SSD then what do you use as swap?

Assuming that you don't (as in most laptops and older desktops) have
the
space for masses of memory?

"Most laptops"? 8G and 16G on the laptops in use here :-)

For some years now I've made sure no system I have anything to do with
is running into swap.


Nice if you can afford the luxury....

Guess how much RAM gridwatch runs on, and how much swap it has?


How much?

384Mbyte

Think there's a gig of swap.

People often think they need more ram than they actually do.

So next question : is TNP's system running into swap? I'd expect it to
not have to. If it's not, then it seems that TNP can indeed "afford the
luxury" of a system not using its swap.

(I do get irritated by eg database software installs which complain that
your swap file is too small. Yes, it's tiny, but that's not a problem
because the server has 60G of memory...)

On 17/10/2016 20:38, Rod Speed wrote:


"Bod" wrote in message
...
On 17/10/2016 19:41, Rod Speed wrote:


"Bod" wrote in message
...
On 17/10/2016 18:32, Rod Speed wrote:


"Bod" wrote in message
...
On 17/10/2016 18:16, Rod Speed wrote:


"Bod" wrote in message
...
On 17/10/2016 16:06, Clive George wrote:
On 17/10/2016 15:12, Bod wrote:
"Afford the luxury"??

If your system is running into swap, it will be slower than it
needs to
be. It's not so much affording the luxury, as affording the
basics.

TNP might be happy running gridwatch on a cheese-pared setup,
but
memory
is really very cheap these days, so anybody doing this sort of
thing for
a living will do it properly - which for most servers means
running in
memory. (there are probably exceptions, I've just not hit them
yet,
but
I do have a fairly wide experience of servers...)

The price depends on what type of memory. 32GB of DDR4 2600 mhz
for my
laptop is £171. I wouldn't call that cheap.

If you're doing this for a living and running something which
needs
32GB
of memory, 171 quid is cheap. You're neither doing it for a
living nor
do you need anything like that, so you think it's expensive.

You probably also think over 500 quid for a laptop is expensive
:-)

(basic reasoning is pretty much that if you're doing it for a
living,
other costs will dwarf it, and if you need it you'll either be
making
rather more than that or saving enough time to make it worth it.)

Has TNP posted how much his gridwatch server has? For
comparison, I
think the cheapest lowest spec Azure VM is 3.5GB, and that will
all be
available as full speed memory if you use it.


My latest laptop cost £900.

So it was stupid to not have more than 8GB of ram
in it and to close apps when you stop using them.

Because that's the amount that was in it when I bought it.
I'm gonna get another 24GB of ram anyway.

So you ****ed your money against the wall on the SSD
and will get no advantage from it once you have 32GB
of ram, dont close apps when you stop using them and
dont stupidly turn the system off every day.

Tell me how fast you can load a pagefull of, say, photos from a
mechanical drive?

Fast enough to not bother to **** any money against
the wall on an SSD. I do that on the smartphone I took
the photos with anyway, not stupid enough to do that
on the desktop or laptop and its instant on the smartphone.


"Hard drives win on price, capacity, and availability. SSDs work best
if speed, ruggedness, form factor, noise, or fragmentation
(technically part of speed) are important factors to you. If it
weren't for the price and capacity issues, SSDs would be the winner
hands down".

http://uk.pcmag.com/storage-devices-...the-difference


Irrelevant to what is being discussed.


You were on about speed. SSDs beat mechanical drives hands down.

"Bod" wrote in message
...
On 17/10/2016 20:38, Rod Speed wrote:


"Bod" wrote in message
...
On 17/10/2016 19:41, Rod Speed wrote:


"Bod" wrote in message
...
On 17/10/2016 18:32, Rod Speed wrote:


"Bod" wrote in message
...
On 17/10/2016 18:16, Rod Speed wrote:


"Bod" wrote in message
...
On 17/10/2016 16:06, Clive George wrote:
On 17/10/2016 15:12, Bod wrote:
"Afford the luxury"??

If your system is running into swap, it will be slower than it
needs to
be. It's not so much affording the luxury, as affording the
basics.

TNP might be happy running gridwatch on a cheese-pared setup,
but
memory
is really very cheap these days, so anybody doing this sort of
thing for
a living will do it properly - which for most servers means
running in
memory. (there are probably exceptions, I've just not hit them
yet,
but
I do have a fairly wide experience of servers...)

The price depends on what type of memory. 32GB of DDR4 2600 mhz
for my
laptop is £171. I wouldn't call that cheap.

If you're doing this for a living and running something which
needs
32GB
of memory, 171 quid is cheap. You're neither doing it for a
living nor
do you need anything like that, so you think it's expensive.

You probably also think over 500 quid for a laptop is expensive
:-)

(basic reasoning is pretty much that if you're doing it for a
living,
other costs will dwarf it, and if you need it you'll either be
making
rather more than that or saving enough time to make it worth it.)

Has TNP posted how much his gridwatch server has? For
comparison, I
think the cheapest lowest spec Azure VM is 3.5GB, and that will
all be
available as full speed memory if you use it.


My latest laptop cost £900.

So it was stupid to not have more than 8GB of ram
in it and to close apps when you stop using them.

Because that's the amount that was in it when I bought it.
I'm gonna get another 24GB of ram anyway.

So you ****ed your money against the wall on the SSD
and will get no advantage from it once you have 32GB
of ram, dont close apps when you stop using them and
dont stupidly turn the system off every day.

Tell me how fast you can load a pagefull of, say, photos from a
mechanical drive?

Fast enough to not bother to **** any money against
the wall on an SSD. I do that on the smartphone I took
the photos with anyway, not stupid enough to do that
on the desktop or laptop and its instant on the smartphone.


"Hard drives win on price, capacity, and availability. SSDs work best
if speed, ruggedness, form factor, noise, or fragmentation
(technically part of speed) are important factors to you. If it
weren't for the price and capacity issues, SSDs would be the winner
hands down".

http://uk.pcmag.com/storage-devices-...the-difference


Irrelevant to what is being discussed.


You were on about speed. SSDs beat mechanical drives hands down.

Not when you have enough of a clue to not close apps when
you have stopped using them for a bit, and arent actually stupid
enough to turn your system off every night and have enough of
a clue to have enough physical memory so it doesnt swap.

On 17/10/2016 21:07, Rod Speed wrote:


"Bod" wrote in message
...
On 17/10/2016 20:38, Rod Speed wrote:


"Bod" wrote in message
...
On 17/10/2016 19:41, Rod Speed wrote:


"Bod" wrote in message
...
On 17/10/2016 18:32, Rod Speed wrote:


"Bod" wrote in message
...
On 17/10/2016 18:16, Rod Speed wrote:


"Bod" wrote in message
...
On 17/10/2016 16:06, Clive George wrote:
On 17/10/2016 15:12, Bod wrote:
"Afford the luxury"??

If your system is running into swap, it will be slower than it
needs to
be. It's not so much affording the luxury, as affording the
basics.

TNP might be happy running gridwatch on a cheese-pared setup,
but
memory
is really very cheap these days, so anybody doing this sort of
thing for
a living will do it properly - which for most servers means
running in
memory. (there are probably exceptions, I've just not hit them
yet,
but
I do have a fairly wide experience of servers...)

The price depends on what type of memory. 32GB of DDR4 2600 mhz
for my
laptop is £171. I wouldn't call that cheap.

If you're doing this for a living and running something which
needs
32GB
of memory, 171 quid is cheap. You're neither doing it for a
living nor
do you need anything like that, so you think it's expensive.

You probably also think over 500 quid for a laptop is expensive
:-)

(basic reasoning is pretty much that if you're doing it for a
living,
other costs will dwarf it, and if you need it you'll either be
making
rather more than that or saving enough time to make it worth
it.)

Has TNP posted how much his gridwatch server has? For
comparison, I
think the cheapest lowest spec Azure VM is 3.5GB, and that will
all be
available as full speed memory if you use it.


My latest laptop cost £900.

So it was stupid to not have more than 8GB of ram
in it and to close apps when you stop using them.

Because that's the amount that was in it when I bought it.
I'm gonna get another 24GB of ram anyway.

So you ****ed your money against the wall on the SSD
and will get no advantage from it once you have 32GB
of ram, dont close apps when you stop using them and
dont stupidly turn the system off every day.

Tell me how fast you can load a pagefull of, say, photos from a
mechanical drive?

Fast enough to not bother to **** any money against
the wall on an SSD. I do that on the smartphone I took
the photos with anyway, not stupid enough to do that
on the desktop or laptop and its instant on the smartphone.


"Hard drives win on price, capacity, and availability. SSDs work best
if speed, ruggedness, form factor, noise, or fragmentation
(technically part of speed) are important factors to you. If it
weren't for the price and capacity issues, SSDs would be the winner
hands down".

http://uk.pcmag.com/storage-devices-...the-difference



Irrelevant to what is being discussed.


You were on about speed. SSDs beat mechanical drives hands down.

Not when you have enough of a clue to not close apps when
you have stopped using them for a bit, and arent actually stupid
enough to turn your system off every night and have enough of
a clue to have enough physical memory so it doesnt swap.

Stupid is your favourite word.

"Bod" wrote in message
...
On 17/10/2016 21:07, Rod Speed wrote:


"Bod" wrote in message
...
On 17/10/2016 20:38, Rod Speed wrote:


"Bod" wrote in message
...
On 17/10/2016 19:41, Rod Speed wrote:


"Bod" wrote in message
...
On 17/10/2016 18:32, Rod Speed wrote:


"Bod" wrote in message
...
On 17/10/2016 18:16, Rod Speed wrote:


"Bod" wrote in message
...
On 17/10/2016 16:06, Clive George wrote:
On 17/10/2016 15:12, Bod wrote:
"Afford the luxury"??

If your system is running into swap, it will be slower than
it
needs to
be. It's not so much affording the luxury, as affording the
basics.

TNP might be happy running gridwatch on a cheese-pared setup,
but
memory
is really very cheap these days, so anybody doing this sort
of
thing for
a living will do it properly - which for most servers means
running in
memory. (there are probably exceptions, I've just not hit
them
yet,
but
I do have a fairly wide experience of servers...)

The price depends on what type of memory. 32GB of DDR4 2600
mhz
for my
laptop is £171. I wouldn't call that cheap.

If you're doing this for a living and running something which
needs
32GB
of memory, 171 quid is cheap. You're neither doing it for a
living nor
do you need anything like that, so you think it's expensive.

You probably also think over 500 quid for a laptop is expensive
:-)

(basic reasoning is pretty much that if you're doing it for a
living,
other costs will dwarf it, and if you need it you'll either be
making
rather more than that or saving enough time to make it worth
it.)

Has TNP posted how much his gridwatch server has? For
comparison, I
think the cheapest lowest spec Azure VM is 3.5GB, and that will
all be
available as full speed memory if you use it.


My latest laptop cost £900.

So it was stupid to not have more than 8GB of ram
in it and to close apps when you stop using them.

Because that's the amount that was in it when I bought it.
I'm gonna get another 24GB of ram anyway.

So you ****ed your money against the wall on the SSD
and will get no advantage from it once you have 32GB
of ram, dont close apps when you stop using them and
dont stupidly turn the system off every day.

Tell me how fast you can load a pagefull of, say, photos from a
mechanical drive?

Fast enough to not bother to **** any money against
the wall on an SSD. I do that on the smartphone I took
the photos with anyway, not stupid enough to do that
on the desktop or laptop and its instant on the smartphone.


"Hard drives win on price, capacity, and availability. SSDs work best
if speed, ruggedness, form factor, noise, or fragmentation
(technically part of speed) are important factors to you. If it
weren't for the price and capacity issues, SSDs would be the winner
hands down".

http://uk.pcmag.com/storage-devices-...the-difference



Irrelevant to what is being discussed.


You were on about speed. SSDs beat mechanical drives hands down.

Not when you have enough of a clue to not close apps when
you have stopped using them for a bit, and arent actually stupid
enough to turn your system off every night and have enough of
a clue to have enough physical memory so it doesnt swap.

Stupid is your favourite word.

I call a spade a spade and stupid stupid.

If you dont like that, stop being stupid.

On 17/10/16 20:07, Bod wrote:
On 17/10/2016 19:46, The Natural Philosopher wrote:
On 17/10/16 14:12, Bod wrote:
On 17/10/2016 13:46, The Natural Philosopher wrote:
On 17/10/16 13:37, Clive George wrote:
On 15/10/2016 15:03, David wrote:
On Sat, 15 Oct 2016 10:51:41 +0100, The Natural Philosopher wrote:

On 15/10/16 10:33, Adrian Caspersz wrote:
snip

Faster booting, faster program load and, if memory is short faster
access to swap if that's on the SSD (although if you want to wear
out an
SSD, using it for swap is one of the best ways).
snip

Just pondering on SSDs and Swap.

If you have a system with just an SSD then what do you use as swap?

Assuming that you don't (as in most laptops and older desktops) have
the
space for masses of memory?

"Most laptops"? 8G and 16G on the laptops in use here :-)

For some years now I've made sure no system I have anything to do with
is running into swap.


Nice if you can afford the luxury....

Guess how much RAM gridwatch runs on, and how much swap it has?


How much?

384Mbyte

Think there's a gig of swap.

People often think they need more ram than they actually do.

I need a gig to really be proof against DOS attacks. I have to actively
block those


--
"Anyone who believes that the laws of physics are mere social
conventions is invited to try transgressing those conventions from the
windows of my apartment. (I live on the twenty-first floor.) "

Alan Sokal

On 17/10/16 20:07, John Rumm wrote:
On 17/10/2016 12:34, The Natural Philosopher wrote:
On 17/10/16 11:31, John Rumm wrote:

It makes far more sense to ensure the partitions are aligned so that the
OS allocation unit is on a 4K boundary (which is the default action on a
modern OS anyway)

My point is that that is what the SSD firmware will in fact do.

No matter where the partition boundary is.

Could you provide a link to a manufacturers documentation for this since
I have not yet seen any SSDs that claim to be able to remap misaligned
partitions?


There are no 'misaligned' partitions.

Just partitions.


--
"Anyone who believes that the laws of physics are mere social
conventions is invited to try transgressing those conventions from the
windows of my apartment. (I live on the twenty-first floor.) "

Alan Sokal

On 17/10/16 20:49, Clive George wrote:
On 17/10/2016 20:07, Bod wrote:
On 17/10/2016 19:46, The Natural Philosopher wrote:
On 17/10/16 14:12, Bod wrote:
On 17/10/2016 13:46, The Natural Philosopher wrote:
On 17/10/16 13:37, Clive George wrote:
On 15/10/2016 15:03, David wrote:
On Sat, 15 Oct 2016 10:51:41 +0100, The Natural Philosopher wrote:

On 15/10/16 10:33, Adrian Caspersz wrote:
snip

Faster booting, faster program load and, if memory is short faster
access to swap if that's on the SSD (although if you want to wear
out an
SSD, using it for swap is one of the best ways).
snip

Just pondering on SSDs and Swap.

If you have a system with just an SSD then what do you use as swap?

Assuming that you don't (as in most laptops and older desktops) have
the
space for masses of memory?

"Most laptops"? 8G and 16G on the laptops in use here :-)

For some years now I've made sure no system I have anything to do
with
is running into swap.


Nice if you can afford the luxury....

Guess how much RAM gridwatch runs on, and how much swap it has?


How much?

384Mbyte

Think there's a gig of swap.

People often think they need more ram than they actually do.

So next question : is TNP's system running into swap? I'd expect it to
not have to. If it's not, then it seems that TNP can indeed "afford the
luxury" of a system not using its swap.

Oh it runs into swap OK especially if its under attack.

Then I have to fight to shut down the webserver and it can take minutes
to get a login prompt

vps:~$ top

top - 21:32:09 up 270 days, 9:44, 1 user, load average: 0.12, 0.16, 0.33
Tasks: 124 total, 2 running, 121 sleeping, 0 stopped, 1 zombie
Cpu(s): 19.2%us, 3.0%sy, 0.0%ni, 53.0%id, 24.5%wa, 0.0%hi, 0.3%si,
0.0%st
Mem: 391680k total, 314636k used, 77044k free, 2804k buffers
Swap: 917488k total, 117092k used, 800396k free, 129112k cached

(I do get irritated by eg database software installs which complain that
your swap file is too small. Yes, it's tiny, but that's not a problem
because the server has 60G of memory...)



--
"Anyone who believes that the laws of physics are mere social
conventions is invited to try transgressing those conventions from the
windows of my apartment. (I live on the twenty-first floor.) "

Alan Sokal

On Mon, 17 Oct 2016 20:20:55 +0100, John Rumm
wrote:

On 17/10/2016 11:42, T i m wrote:
On Mon, 17 Oct 2016 11:31:48 +0100, John Rumm
wrote:

On 17/10/2016 10:45, The Natural Philosopher wrote:
On 17/10/16 09:37, John Rumm wrote:
In reality it could be anywhere on any boundary the
SSD firmware and processor puts it.

Not quite - there is not a one to one mapping of OS allocation units to
flash pages. For optimum performance you need to ensure that whichever
allocation unit you update, the SSD can do that update by operating on
one (and only one) page of flash. With the wrong alignment, you can end
up with the SSD needing to do two page updates for each OS allocation
unit update.


However all that may be true, but its not under user level control via
partioning, and its handled internally by the SSD.

There is no obvious way a SSD could make a sensible choice to internally
remap alignment if it turns out you have managed to install an OS
partition with a start LBA offset from the ideal. Especially as one
physical drive can host more than one partition, and if you really
tried, you could end up with several partitions each with different
alignments.

It makes far more sense to ensure the partitions are aligned so that the
OS allocation unit is on a 4K boundary (which is the default action on a
modern OS anyway)

So and irrespective of any performance impact ITRW, if some software
(Gparted) can see and display that these alignments aren't made *and*
can set them, is Gparted actually then *actually / physically*
resetting said alignments or just indicating it is?

It would need to copy the entire partition up or down a number of
sectors.

Yes, I have seen how one used Gparted to do it John but was interested
to know how we *know* it was actually doing it? Like when your tried
to LL format a drive and it looked like it was doing so but then you
rebooted and all your stuff was still there ... eg, Is it actually
happening (and how can we *prove* that) or is it simply saying it's
happening (by say moving a pointer) but not *actually* doing it (as
TNP seem to suggest is the case)?

Or like when you could no longer set the right drive geometry up
(c,h,s/t, b/s) and so the drives started doing translation to allow
you to access the whole drive past whatever BOIS limit was in place at
the time?

(However the OS may bork and need a reinstall if you do!)

Quite. ;-(

What would be a good (valid) way of checking for such things
(increased performance hopefully) pre and post adjustment?

On windows, do a msinfo32 in a command prompt to start the system
information tool. Then expand out the Components - Storage - Disks
part of the tree view. That will show all the physical drives, and the
partitions and their starting offsets. Divide the offset by 4096 and
hope you get an integer answer ;-)

e.g:

Description Disk drive
Manufacturer (Standard disk drives)
Model KINGSTON SHSS37A240G ATA Device
Bytes/Sector 512
Media Loaded Yes
Media Type Fixed hard disk
Partitions 1
SCSI Bus 0
SCSI Logical Unit 0
SCSI Port 0
SCSI Target ID 0
Sectors/Track 63
Size 223.57 GB (240,054,796,800 bytes)
Total Cylinders 29,185
Total Sectors 468,857,025
Total Tracks 7,442,175
Tracks/Cylinder 255
Partition Disk #0, Partition #0
Partition Size 223.57 GB (240,055,746,560 bytes)
Partition Starting Offset 1,048,576 bytes

Ok, but again, how do we *know* we are seeing the raw data and not
some soft translation of same? Is there some diagnostic software that
can really access the drive and report what *it* is actually doing?

It all sounds like re-numbering sector to reduce latency with Optune
all those years ago. ;-)

Yup, when disc controllers were so slow they might not be able to read
consecutive sectors if they were physically adjacent to each other!

That was it ... and testing it and resetting it was like re-plastering
the wall without touching the wallpaper. ;-)

Cheers, T i m

On Mon, 17 Oct 2016 21:18:18 +0100, pamela wrote:

On 11:42 17 Oct 2016, T i m wrote:

On Mon, 17 Oct 2016 11:31:48 +0100, John Rumm
wrote:

On 17/10/2016 10:45, The Natural Philosopher wrote:
On 17/10/16 09:37, John Rumm wrote:
In reality it could be anywhere on any boundary the
SSD firmware and processor puts it.

Not quite - there is not a one to one mapping of OS
allocation units to flash pages. For optimum performance you
need to ensure that whichever allocation unit you update, the
SSD can do that update by operating on one (and only one)
page of flash. With the wrong alignment, you can end up with
the SSD needing to do two page updates for each OS allocation
unit update.


However all that may be true, but its not under user level
control via partioning, and its handled internally by the SSD.

There is no obvious way a SSD could make a sensible choice to
internally remap alignment if it turns out you have managed to
install an OS partition with a start LBA offset from the ideal.
Especially as one physical drive can host more than one
partition, and if you really tried, you could end up with
several partitions each with different alignments.

It makes far more sense to ensure the partitions are aligned so
that the OS allocation unit is on a 4K boundary (which is the
default action on a modern OS anyway)

So and irrespective of any performance impact ITRW, if some
software (Gparted) can see and display that these alignments
aren't made *and* can set them, is Gparted actually then
*actually / physically* resetting said alignments or just
indicating it is?

What would be a good (valid) way of checking for such things
(increased performance hopefully) pre and post adjustment?

It all sounds like re-numbering sector to reduce latency with
Optune all those years ago. ;-)

Cheers, T i m

You can still buy Spin Rite which used to do that sector
interleaving.

Hmmm, I though I remembered using Optune for that but now you have me
thinking it could have been SpinRite. I'm pretty sure I used both at
the time though, along with NDD and NC.

http://www.danielsays.com/ss-gallery-dos-optune-12.htm

This looks more familiar though ...

https://en.wikipedia.org/wiki/SpinRi...inrite-2.0.png


It now has other bells and whistles, some of
disputed benefit, for a mere $89!

Bargain. ;-)

Cheers, T i m

On 15/10/2016 10:11, Peter Johnson wrote:
On Sat, 15 Oct 2016 09:54:16 +0100, Bod wrote:



Hmm! that hasn't been my experience. I've changed several laptops to
SSDs and *everything* is much snappier.

I put an SSD into a netbook and am surprised at the difference it
made.

Made an amazing difference in an iMac. Like a new machine!

--
Rod

En el artículo , pamela
escribió:

He sees the SSD presenting itself through a logical representation
behind which all essential activities take place hidden and without
any user influence.

It does, but not to the extent of recognising partitions are mis-aligned
and fixing it. For that to happen, the firmware on the SSD would have
to be partition table and filesystem aware, one level up from being a
block device (looking to the host machine like a box of LBAs labelled
from 0 to n, where n is the last available block.)

I see he's getting a beating on another thread about, of all things,
rocket science.

--
(\_/)
(='.'=) systemd: the Linux version of Windows 10
(")_(")

On 17/10/2016 22:45, polygonum wrote:
On 15/10/2016 10:11, Peter Johnson wrote:
On Sat, 15 Oct 2016 09:54:16 +0100, Bod wrote:



Hmm! that hasn't been my experience. I've changed several laptops to
SSDs and *everything* is much snappier.

I put an SSD into a netbook and am surprised at the difference it
made.

Made an amazing difference in an iMac. Like a new machine!

Yes it can make that sort of dramatic difference.
Rod still won't believe it though ;-)

"Bod" wrote in message
...
On 17/10/2016 22:45, polygonum wrote:
On 15/10/2016 10:11, Peter Johnson wrote:
On Sat, 15 Oct 2016 09:54:16 +0100, Bod wrote:



Hmm! that hasn't been my experience. I've changed several laptops to
SSDs and *everything* is much snappier.

I put an SSD into a netbook and am surprised at the difference it
made.

Made an amazing difference in an iMac. Like a new machine!

Yes it can make that sort of dramatic difference.

Only if you use it stupidly or stupidly don’t have
enough physical memory to stop it swapping.

On 18/10/2016 07:48, Rod Speed wrote:


"Bod" wrote in message
...
On 17/10/2016 22:45, polygonum wrote:
On 15/10/2016 10:11, Peter Johnson wrote:
On Sat, 15 Oct 2016 09:54:16 +0100, Bod wrote:



Hmm! that hasn't been my experience. I've changed several laptops to
SSDs and *everything* is much snappier.

I put an SSD into a netbook and am surprised at the difference it
made.

Made an amazing difference in an iMac. Like a new machine!

Yes it can make that sort of dramatic difference.

Only if you use it stupidly or stupidly don’t have
enough physical memory to stop it swapping.


I've already told you that I don't use nor need a swap file.

"Bod" wrote in message
...
On 18/10/2016 07:48, Rod Speed wrote:


"Bod" wrote in message
...
On 17/10/2016 22:45, polygonum wrote:
On 15/10/2016 10:11, Peter Johnson wrote:
On Sat, 15 Oct 2016 09:54:16 +0100, Bod wrote:



Hmm! that hasn't been my experience. I've changed several laptops to
SSDs and *everything* is much snappier.

I put an SSD into a netbook and am surprised at the difference it
made.

Made an amazing difference in an iMac. Like a new machine!

Yes it can make that sort of dramatic difference.

Only if you use it stupidly or stupidly don’t have
enough physical memory to stop it swapping.


I've already told you that I don't use nor need a swap file.

Irrelevant to what the other Rod does with that imac.

On Mon, 17 Oct 2016 20:20:55 +0100, John Rumm
wrote:
snip

What would be a good (valid) way of checking for such things
(increased performance hopefully) pre and post adjustment?

On windows, do a msinfo32 in a command prompt to start the system
information tool. Then expand out the Components - Storage - Disks
part of the tree view. That will show all the physical drives, and the
partitions and their starting offsets. Divide the offset by 4096 and
hope you get an integer answer ;-)

e.g:

snip

Partition Starting Offset 1,048,576 bytes

So, on my best laptop (Tosh A300) with a 240G SSD it looks like my
starting offsets are all out (W10 System, Data, Mint root, Mint Home,
Mint swap).

So, how could I test the performance pre and post re-alignment please,
so we can see just how much difference it *really* makes (if any)?

Cheers, T i m

On 18/10/16 07:11, Mike Tomlinson wrote:
En el artÃ*culo , pamela
escribió:

He sees the SSD presenting itself through a logical representation
behind which all essential activities take place hidden and without
any user influence.

It does, but not to the extent of recognising partitions are mis-aligned
and fixing it. For that to happen, the firmware on the SSD would have
to be partition table and filesystem aware,

And you think it isn't?

one level up from being a
block device (looking to the host machine like a box of LBAs labelled
from 0 to n, where n is the last available block.)

I see he's getting a beating on another thread about, of all things,
rocket science.



--
"What do you think about Gay Marriage?"
"I don't."
"Don't what?"
"Think about Gay Marriage."

On 17/10/2016 22:15, T i m wrote:
On Mon, 17 Oct 2016 20:20:55 +0100, John Rumm
wrote:

On 17/10/2016 11:42, T i m wrote:

So and irrespective of any performance impact ITRW, if some software
(Gparted) can see and display that these alignments aren't made *and*
can set them, is Gparted actually then *actually / physically*
resetting said alignments or just indicating it is?

It would need to copy the entire partition up or down a number of
sectors.

Yes, I have seen how one used Gparted to do it John but was interested
to know how we *know* it was actually doing it?

That's basically saying "does Gparted's partition resize / move
capability work?" The answer to which is yes.

Like when your tried
to LL format a drive and it looked like it was doing so but then you
rebooted and all your stuff was still there ... eg,

Low level formatting on IDE onwards was "supported" by issuing a command
to the drive and letting it get on with it (or not). Moving a partition
is not something a drive has embedded support for - your external
utility needs to physically read a bunch of blocks from one area of the
disk and write them back somewhere else, then rinse and repeat until its
done the whole partition.

Is it actually
happening (and how can we *prove* that) or is it simply saying it's
happening (by say moving a pointer) but not *actually* doing it (as
TNP seem to suggest is the case)?

Its happening for real, the partition will appear at a new starting
logical block address once done.

Internally the shift in content may not be reflected in exactly the same
way as a result of wear levelling, but its also fair to say the drive is
not going to be smart enough to say "aha, that 512MB bunch of sectors
you read a few mins ago, looks suspiciously similar to a new bunch of
sectors you are now writing 600 meg higher up the LBA address space -
you know what, I will take a gamble that you won't be needing the
original copy of that data and just pull a fast one swapping some
pointers round"


What would be a good (valid) way of checking for such things
(increased performance hopefully) pre and post adjustment?

On windows, do a msinfo32 in a command prompt to start the system
information tool. Then expand out the Components - Storage - Disks
part of the tree view. That will show all the physical drives, and the
partitions and their starting offsets. Divide the offset by 4096 and
hope you get an integer answer ;-)

Ok, but again, how do we *know* we are seeing the raw data and not
some soft translation of same?

You are seeing the raw partition info and LBA numbers.

Even with a normal HDD there may be sector remapping going on, so you
won't necessarily know how those LBAs map to the physical CHS addresses
on the HDD, or the flash page address of the SSD (especially as SSDs use
internal parallelism for performance)

Is there some diagnostic software that
can really access the drive and report what *it* is actually doing?

With SSD you probably can't get at physical flash pages unless you have
a proprietary tool from the maker of the drive. (one of the reasons why
there is a security risk when attempting to sanitise a SSD - you can't
be sure you have got all of the flash pages since they are not all
visible via a LBA - and the physical page used for a given LBA may
change in use).



--
Cheers,

John.

/================================================== ===============\
| Internode Ltd - http://www.internode.co.uk |
|-----------------------------------------------------------------|
| John Rumm - john(at)internode(dot)co(dot)uk |
\================================================= ================/

On 17/10/2016 21:30, The Natural Philosopher wrote:
On 17/10/16 20:07, John Rumm wrote:
On 17/10/2016 12:34, The Natural Philosopher wrote:
On 17/10/16 11:31, John Rumm wrote:

It makes far more sense to ensure the partitions are aligned so that
the
OS allocation unit is on a 4K boundary (which is the default action
on a
modern OS anyway)

My point is that that is what the SSD firmware will in fact do.

No matter where the partition boundary is.

Could you provide a link to a manufacturers documentation for this since
I have not yet seen any SSDs that claim to be able to remap misaligned
partitions?


There are no 'misaligned' partitions.

Just partitions.

OK, so if you look at the Intel document from 2014[1], they explain why
misaligned partitions cause performance impacts on their drives of that
time. You have claimed that that information is out of date, and its no
longer an issue. Could you point to an authoritative source that
justifies that claim?

Or are you claiming that Intel drives are just way behind the current
state of the art, and other makers drives don't suffer the problem, but
then Intel ones still do? In which case it sounds like a strong
marketing hook the other drive makers would use. Could you point to
some documentation reflecting this?


[1]
http://www.intel.co.uk/content/dam/w...tech-brief.pdf


--
Cheers,

John.

/================================================== ===============\
| Internode Ltd - http://www.internode.co.uk |
|-----------------------------------------------------------------|
| John Rumm - john(at)internode(dot)co(dot)uk |
\================================================= ================/

On 18/10/16 13:24, John Rumm wrote:
On 17/10/2016 21:30, The Natural Philosopher wrote:
On 17/10/16 20:07, John Rumm wrote:
On 17/10/2016 12:34, The Natural Philosopher wrote:
On 17/10/16 11:31, John Rumm wrote:

It makes far more sense to ensure the partitions are aligned so that
the
OS allocation unit is on a 4K boundary (which is the default action
on a
modern OS anyway)

My point is that that is what the SSD firmware will in fact do.

No matter where the partition boundary is.

Could you provide a link to a manufacturers documentation for this since
I have not yet seen any SSDs that claim to be able to remap misaligned
partitions?


There are no 'misaligned' partitions.

Just partitions.

OK, so if you look at the Intel document from 2014[1], they explain why
misaligned partitions cause performance impacts on their drives of that
time. You have claimed that that information is out of date, and its no
longer an issue. Could you point to an authoritative source that
justifies that claim?

Or are you claiming that Intel drives are just way behind the current
state of the art, and other makers drives don't suffer the problem, but
then Intel ones still do? In which case it sounds like a strong
marketing hook the other drive makers would use. Could you point to
some documentation reflecting this?


If you look at the actual strategies - and I supplied pointers to then -
google 'wear levelling' you will realise that is a massively complex
process right away, and completely destroys any relationship between
what the disk reports as a block device and where physically the data is
stored. Yea even unto splitting blocks of data up into different flash
pages or amalgamating two different 'sectors'; into one flash page.

IN essence the disk is a database, with sectors located wherever they
happen to end up, and indexed by a database like structure that relates
the sector number of the logical device, to an offset in a flash page
somewhere on the disk.

And tuning the garbage correction and read and write algorithms is the
best way to increase the system life.

Logical sectors that have never been written , simply do not have any
physical area assigned to them.

The asertion of the partitioneers, is that if you start a partition with
whatever information it has in it, on a less rthan e.g. 4K boundary,
that will result in more cross-page writes. Firstly I dont see that is
true even if the algorithm is always to apply aligned sectors in 4k
chunks to a 4k flash page.

But I see no reason why if the disk detects a sequence of data starting
at say 133rd sector....it would not start that at the beginning of the
flash page, assuming it has no prior knowledge of this sequence.

After all, all that is being maintained is a database of Logical
sector-physical flash pageage offset.

There is no reason to say that because the sector number is, modulo 4,
number 3, that the page offset should also be 3.

Note that nearly all of this IIRC only applies to windows, as Linux file
systems are aligned on much larger page sizes anyway.


The point I wanted to make way back before all the name calling started,
was that they way data is physically organised on a flash disk bears no
relation whatsoever to how it is presented logically. It is organised
precisely to minimise writes and erases, and to ensure that when those
take place, they are as evenly spread as possible.

Therefore theories that changing the nature of OS access to the
*logical* structure will impact on the physical processes are at best
optimistic simplistic and dependent on a poor implementation of wear
levelling and garbage collection algorithms. At worst they are risible
nonsense.

And there is evidence that more than one manufacturer has firmware that
exactly understands partitions of the NTFS sort and has adjusted its
algorithms accordingly. Why would they not?

https://en.wikipedia.org/wiki/Write_...age_collection


A little knowledge is a dangerous thing., To someone with a little
knowledge, someone with a lot more appears ignorant and uninformed.

The 'partition your SSD correctly' appears to be an urban (windows)
myth, that has propagated widely, and yet no explanation of why it works
is consistent with my understanding of how logical to physical mapping
works in a real life SSD.

And because I have written code to analyse and understand file systems
at the sector level, I know its really not hard to understand what type
of FS and what structure is located where, on a disk.

The crucial step is of course coding that can cope with arbitrary
locations of a 512 byte sector, rather than a 4k or 16k block that it
might be part of.

I can conceive that a rubbish programmer with no understanding of file
systems or flash might cobble up something that bad for a prototype
product, but in a fiercely competitive field with longevity being an
essential part of the product attributes, its unlikely that better
algorithms and more flexible mapping systems would not evolve.



[1]
http://www.intel.co.uk/content/dam/w...tech-brief.pdf



Is an interesting document to be sure.

I cam also point you at documentation by quite large companies showing
how to reduce your carbon emissions with windmills, and how seroxat is a
wonder drug with no side effects.

My best guess is that someone challenged Intel with some urban myth, and
some nerd got tasked to write some 'support literature' to 'take care of
it' from a marketing POV, the technical solution already having moved on...

But even if its in fact true, it is so for only one make of drive at one
particular time. As I said. tuning the firmware is big business and
means lots of dollars. There are half a dozen or more people who own and
sell firmware and SSD controller chipsets to manage SSDS.

Who knows WHAT goes in inside?

Pretty much only the people who wrote the firmware, in detail BUT the
flow of product information and PHd papers shows which way the wind is
blowing an the general classes of solutions. See the wiki article on
wear levelling.

PS for those paranoid about security, note that writing zeros to every
sector on an SSD disk does not begin to guarantee the erasure of ANY
data on the chip at all.

Now should we talk about internal data compression...;-)?






--
Microsoft : the best reason to go to Linux that ever existed.

On Mon, 17 Oct 2016 07:03:27 +0100, Rod Speed wrote:



"James Wilkinson Sword" wrote in message
news
On Sat, 15 Oct 2016 09:54:16 +0100, Bod wrote:

On 15/10/2016 09:44, alan_m wrote:
On 15/10/2016 01:23, David Paste wrote:
am I barking up the
wrong tree to assume an SSD will improve the snappiness of the user
interface?)


When I replaced a traditional hard disk in my 5 year old laptop with a
SSD I noticed a faster start up but for day to day use no overall
difference in speed.




Hmm! that hasn't been my experience. I've changed several laptops to
SSDs and *everything* is much snappier.

Agreed. I guess his day to day use is opening one word processor and
using it for hours.

Nope, just only rebooting ever few months and not closing apps.

A very odd way to use a computer.

--
In 2005 eight Brits (All Scottish) cracked their skulls while throwing up into the toilet.