Greetings
We're looking to upgrade two sites which consist of a FAS2020 (7.3.7) and FAS2240-2 (8.1.4-7m) at each site. The 2240-2 systems have a FlashPool accelerated aggregate and the customer is happy with the performance.
The new filers will be either a FAS2552 with additional shelves or we may ditch the entry toasters and go with a FAS8020 system and shelves at each site. If we go with the 2552 we'll keep using FlashPool. If we go with the 8020 we have the option of going FlashCache.
We'd like to evaluate the potential performance improvements of using FlashCache on the 8020 in order to help with the purchasing decision. However, reading TR-3801 it looks like the FAS2000 and 2200 platforms don't support Predictive Cache Statistics (PCS). This is confirmed by trying to enable PCS:
toaster> options flexscale.enable on FlexScale PCS is not active and cannot be enabled.
Are there any other tools which can help us estimate the performance benefits of FlashCache? Or perhaps some key metrics we can collect and analyse?
Cheers Steve
-------------------------------------------------------------------------------- This e-mail message has been scanned for Viruses and Content and cleared by MailMarshal --------------------------------------------------------------------------------
disclaimer
This email communication does not create or vary any contractual relationship between LogicalisCI and you. Internet communications are not secure and accordingly LogicalisCI does not accept any legal liability for the contents of this message. The contents of this email are confidential to the intended recipient at the email address to which it has been addressed. It may not be disclosed to or used by anyone other than this addressee, nor may it be copied in any way. If received in error, please contact LogicalisCI on the above number (listed as Main) quoting the name of the sender and the addressee and then delete it from your system. Please note that neither LogicalisCI nor the sender accepts any responsibility for viruses and it is your responsibility to scan the email and attachments (if any).
Please be aware that LogicalisCI may monitor email traffic data and also email content for security purposes.
The following companies are collectively referred to in the above statements as LogicalisCI: Logicalis Channel Islands Limited, Registered in Jersey No: 67161, Registered Office: Rue A La Dame, Five Oaks, St. Saviour, Jersey, JE2 7NH Logicalis Jersey Limited, Registered in Jersey No: 10124, Registered Office: Rue A La Dame, Five Oaks, St. Saviour, Jersey, JE2 7NH Logicalis Guernsey Limited, Registered in Guernsey No: 10896, Registered Office: Pitronnerie Road, St. Peter Port, Guernsey, GY1 2RF
Stephen Stocke wrote:
We're looking to upgrade two sites which consist of a FAS2020 (7.3.7) and FAS2240-2 (8.1.4-7m) at each site. The 2240-2 systems have a FlashPool accelerated aggregate and the customer is happy with the performance.
So these small 2240-2 boxes, have how much FlashPool each...? One Aggr I assume?
The new filers will be either a FAS2552 with additional shelves or we may ditch the entry toasters and go with a FAS8020 system and shelves at each site. If we go with the 2552 we’ll keep using FlashPool. If we go with the 8020 we have the option of going FlashCache.
So your choice is A) a new FAS2252 with add shelves and FlashPool, or B) a new FAS8020 with FlashPool or FlashCache
is that it? If everyone's good with how FlashPool works in the older/smaller 2240-2 systems you kind of have a basline I'd say. Why would you want to switch over to FlashCache at all? It won't improve anything, and it's not cheaper. The only difference before 8.3.1 (which is out in RC1 now) is that w FlashPool you cannot tune things quite as much as you can in FlashCache since a very long time now (lopri_blocks, large seq R benefits from this):
$ naXXX options flexscale flexscale.enable on (same value in local+partner recommended) flexscale.lopri_blocks on (same value in local+partner recommended) flexscale.normal_data_blocks on (same value in local+partner recommended) flexscale.readahead_blocks off (same value in local+partner recommended) flexscale.rewarm on (same value in local+partner recommended)
It's really not trivial to say neither how much cache you should have, nor if you're better off with lopri_blocks off or on. It is so very dependend on the workload parameters (temporal locality and working set size as well as the mix of ops: random/seq). Personally I feel that it's much easier to understand how FlashCache works, than what FlashPool is doing. Very very different -- vastly dissimmilar places in the whole system so to speak.
For FlashCache you watch the "churn" through the cache and then it's just the Hit Rate and Disk Reads Replaced that matters (higher = better)
Your customers are happy with the performance of the entry level (tiny, from my PoV) 2240 and 2252, so I think this is sub-optimisation at a higher level ;-)
Stephen Stocke wrote:
We'd like to evaluate the potential performance improvements of using FlashCache on the 8020 in order to help with the purchasing decision.
There's likely no difference for you in an 8020 between FlashPool and FlashCache. Especially with 8.3.1 where the -Pool is more tunable than before [see above]
I'd just continue to use FlashPool since you're already there now. No point in starting to use FlashCache too, -Pool is the way fwd in any case and an FAS8020 can have more and more flexible configs of -Pool than -Cache, so...
/M
For FlashPool you can set caching policy per volume:
• none - Does not cache any user data or metadata blocks. • auto - Read caches all metadata and randomly read user data blocks and write caches all randomly overwritten user data blocks. • meta - Read caches only metadata blocks. • random_read - Read caches all metadata and randomly read user data blocks. • random_read_write - Read caches all metadata, randomly read and randomly written user data blocks. • meta-random_write - Read caches all metadata and write caches randomly overwritten user data blocks. • noread-random_write - Write caches all randomly overwritten user data blocks. It does not do any read caching. • random_read_write-random-write - Read caches all metadata, randomly read and randomly written user data blocks. It also write caches randomly overwritten user data blocks.
-----Original Message----- From: toasters-bounces@teaparty.net [mailto:toasters-bounces@teaparty.net] On Behalf Of Michael Bergman Sent: Tuesday, June 30, 2015 5:00 PM To: Toasters Subject: Re: FAS2xxx Upgrade - Flash Cache Performance Estimation?
Stephen Stocke wrote:
We're looking to upgrade two sites which consist of a FAS2020 (7.3.7) and FAS2240-2 (8.1.4-7m) at each site. The 2240-2 systems have a FlashPool accelerated aggregate and the customer is happy with the performance.
So these small 2240-2 boxes, have how much FlashPool each...? One Aggr I assume?
The new filers will be either a FAS2552 with additional shelves or we may ditch the entry toasters and go with a FAS8020 system and shelves at each site. If we go with the 2552 we’ll keep using FlashPool. If we go with the 8020 we have the option of going FlashCache.
So your choice is A) a new FAS2252 with add shelves and FlashPool, or B) a new FAS8020 with FlashPool or FlashCache
is that it? If everyone's good with how FlashPool works in the older/smaller 2240-2 systems you kind of have a basline I'd say. Why would you want to switch over to FlashCache at all? It won't improve anything, and it's not cheaper. The only difference before 8.3.1 (which is out in RC1 now) is that w FlashPool you cannot tune things quite as much as you can in FlashCache since a very long time now (lopri_blocks, large seq R benefits from this):
$ naXXX options flexscale flexscale.enable on (same value in local+partner recommended) flexscale.lopri_blocks on (same value in local+partner recommended) flexscale.normal_data_blocks on (same value in local+partner recommended) flexscale.readahead_blocks off (same value in local+partner recommended) flexscale.rewarm on (same value in local+partner recommended)
It's really not trivial to say neither how much cache you should have, nor if you're better off with lopri_blocks off or on. It is so very dependend on the workload parameters (temporal locality and working set size as well as the mix of ops: random/seq). Personally I feel that it's much easier to understand how FlashCache works, than what FlashPool is doing. Very very different -- vastly dissimmilar places in the whole system so to speak.
For FlashCache you watch the "churn" through the cache and then it's just the Hit Rate and Disk Reads Replaced that matters (higher = better)
Your customers are happy with the performance of the entry level (tiny, from my PoV) 2240 and 2252, so I think this is sub-optimisation at a higher level ;-)
Stephen Stocke wrote:
We'd like to evaluate the potential performance improvements of using FlashCache on the 8020 in order to help with the purchasing decision.
There's likely no difference for you in an 8020 between FlashPool and FlashCache. Especially with 8.3.1 where the -Pool is more tunable than before [see above]
I'd just continue to use FlashPool since you're already there now. No point in starting to use FlashCache too, -Pool is the way fwd in any case and an FAS8020 can have more and more flexible configs of -Pool than -Cache, so...
/M
_______________________________________________ Toasters mailing list Toasters@teaparty.net http://www.teaparty.net/mailman/listinfo/toasters
Andrei, Michael, thanks for the prompt replies and input.
To answer Michael's questions:
So these small 2240-2 boxes, have how much FlashPool each...? One Aggr I assume?
In addition to the small aggr hosting vol0, the 2240 boxes have one SAS+SSD FP aggr and one SATA aggr. These aggregates are split between the controllers. The system has 4 x 200GB SSD (2 data disks in the SSD RG). In DOT 8.1 that's the max FlashPool size for the entire system on the 2200 platform. When we upgrade it to 8.2.x the max will increase to 800GB which is still rather small.
So your choice is A) a new FAS2252 with add shelves and FlashPool, or B) a new FAS8020 with FlashPool or FlashCache
is that it?
That pretty well sums it up. The main requirement is to the replace the 2020 systems as they are nearing EOL and the DS14MK2-AT shelves are also nearing EOL. We could add up to 4 more shelves to the 2240 but the customer wants to expand their use of flash acceleration and we've reached the max FlashPool on the 2240.
If everyone's good with how FlashPool works in the older/smaller 2240-2 systems you kind of have a basline I'd say. Why >would you want to switch over to FlashCache at all? It won't improve anything, and it's not cheaper.
We don't necessarily want to go FlashCache, we simply want to choose the right solution for the workload. My main issue is that, with these platforms, there doesn't appear to be a good way to evaluate the right amount of Flash Cache/Pool required when sizing the upgrade.
Perhaps I should have phrased the orignal question differently...
When planning an upgrade for a FAS2000 series filer, how do you size the required FlashCache or FlashPool storage for the workload without having PCS or AWA available?
Steve
-----Original Message----- From: toasters-bounces@teaparty.net [mailto:toasters-bounces@teaparty.net] On Behalf Of Borzenkov, Andrei Sent: 30 June 2015 15:16 To: Michael Bergman; Toasters Subject: RE: FAS2xxx Upgrade - Flash Cache Performance Estimation?
For FlashPool you can set caching policy per volume:
* none - Does not cache any user data or metadata blocks. * auto - Read caches all metadata and randomly read user data blocks and write caches all randomly overwritten user data blocks. * meta - Read caches only metadata blocks. * random_read - Read caches all metadata and randomly read user data blocks. * random_read_write - Read caches all metadata, randomly read and randomly written user data blocks. * meta-random_write - Read caches all metadata and write caches randomly overwritten user data blocks. * noread-random_write - Write caches all randomly overwritten user data blocks. It does not do any read caching. * random_read_write-random-write - Read caches all metadata, randomly read and randomly written user data blocks. It also write caches randomly overwritten user data blocks.
-----Original Message----- From: toasters-bounces@teaparty.net [mailto:toasters-bounces@teaparty.net] On Behalf Of Michael Bergman Sent: Tuesday, June 30, 2015 5:00 PM To: Toasters Subject: Re: FAS2xxx Upgrade - Flash Cache Performance Estimation?
Stephen Stocke wrote:
We're looking to upgrade two sites which consist of a FAS2020 (7.3.7) and FAS2240-2 (8.1.4-7m) at each site. The 2240-2 systems have a FlashPool accelerated aggregate and the customer is happy with the performance.
So these small 2240-2 boxes, have how much FlashPool each...? One Aggr I assume?
The new filers will be either a FAS2552 with additional shelves or we may ditch the entry toasters and go with a FAS8020 system and shelves at each site. If we go with the 2552 we'll keep using FlashPool. If we go with the 8020 we have the option of going FlashCache.
So your choice is A) a new FAS2252 with add shelves and FlashPool, or B) a new FAS8020 with FlashPool or FlashCache
is that it? If everyone's good with how FlashPool works in the older/smaller 2240-2 systems you kind of have a basline I'd say. Why would you want to switch over to FlashCache at all? It won't improve anything, and it's not cheaper. The only difference before 8.3.1 (which is out in RC1 now) is that w FlashPool you cannot tune things quite as much as you can in FlashCache since a very long time now (lopri_blocks, large seq R benefits from this):
$ naXXX options flexscale flexscale.enable on (same value in local+partner recommended) flexscale.lopri_blocks on (same value in local+partner recommended) flexscale.normal_data_blocks on (same value in local+partner recommended) flexscale.readahead_blocks off (same value in local+partner recommended) flexscale.rewarm on (same value in local+partner recommended)
It's really not trivial to say neither how much cache you should have, nor if you're better off with lopri_blocks off or on. It is so very dependend on the workload parameters (temporal locality and working set size as well as the mix of ops: random/seq). Personally I feel that it's much easier to understand how FlashCache works, than what FlashPool is doing. Very very different -- vastly dissimmilar places in the whole system so to speak.
For FlashCache you watch the "churn" through the cache and then it's just the Hit Rate and Disk Reads Replaced that matters (higher = better)
Your customers are happy with the performance of the entry level (tiny, from my PoV) 2240 and 2252, so I think this is sub-optimisation at a higher level ;-)
Stephen Stocke wrote:
We'd like to evaluate the potential performance improvements of using FlashCache on the 8020 in order to help with the purchasing decision.
There's likely no difference for you in an 8020 between FlashPool and FlashCache. Especially with 8.3.1 where the -Pool is more tunable than before [see above]
I'd just continue to use FlashPool since you're already there now. No point in starting to use FlashCache too, -Pool is the way fwd in any case and an FAS8020 can have more and more flexible configs of -Pool than -Cache, so...
/M
_______________________________________________ Toasters mailing list Toasters@teaparty.net http://www.teaparty.net/mailman/listinfo/toasters
_______________________________________________ Toasters mailing list Toasters@teaparty.net http://www.teaparty.net/mailman/listinfo/toasters -------------------------------------------------------------------------------- This e-mail message has been scanned for Viruses and Content and cleared by MailMarshal --------------------------------------------------------------------------------
disclaimer
This email communication does not create or vary any contractual relationship between LogicalisCI and you. Internet communications are not secure and accordingly LogicalisCI does not accept any legal liability for the contents of this message. The contents of this email are confidential to the intended recipient at the email address to which it has been addressed. It may not be disclosed to or used by anyone other than this addressee, nor may it be copied in any way. If received in error, please contact LogicalisCI on the above number (listed as Main) quoting the name of the sender and the addressee and then delete it from your system. Please note that neither LogicalisCI nor the sender accepts any responsibility for viruses and it is your responsibility to scan the email and attachments (if any).
Please be aware that LogicalisCI may monitor email traffic data and also email content for security purposes.
The following companies are collectively referred to in the above statements as LogicalisCI: Logicalis Channel Islands Limited, Registered in Jersey No: 67161, Registered Office: Rue A La Dame, Five Oaks, St. Saviour, Jersey, JE2 7NH Logicalis Jersey Limited, Registered in Jersey No: 10124, Registered Office: Rue A La Dame, Five Oaks, St. Saviour, Jersey, JE2 7NH Logicalis Guernsey Limited, Registered in Guernsey No: 10896, Registered Office: Pitronnerie Road, St. Peter Port, Guernsey, GY1 2RF
Stephen -
If you work with your reseller they have access to modeling tools that can help.
I really like the 8020 but keep in mind the # of PCIe slots is very limited (2 per controller) so if you were to add in flash cache cards you'll only have one slot left. Most people order their 8020 with an extra SAS card so they can run at least two stacks (maybe a SAS stack and a SATA stack) and that last slot usually goes to a 10 Gb card - remember there's only 4 10Gb ports per controller and 2 are immediately consumed by your cluster network, leaving two for data which may or may not be enough for your environment. The short answer is that most of the 8020s I see in the field are using Flash Pools and not Flash Cache due to the limited expansion options.
If you really want Flash Cache I would suggest working with your reseller to get a good deal on an 8040. With the 8040 you basically have enough SAS ports built in to support two stacks and enough onboard 10Gb/UTA2 ports for flexible connectivity options, so those PCIe slots can be used for Flash Cache without feeling like you're boxed in.
Don't get me wrong, the 8020 is a great little box, but might not be exactly what you're looking for in this particular instance, especially if you're set on using Flash Cache.
On Jun 30, 2015, at 5:56 AM, Stephen Stocke <Stephen.Stocke@je.logicalis.commailto:Stephen.Stocke@je.logicalis.com> wrote:
Greetings
We’re looking to upgrade two sites which consist of a FAS2020 (7.3.7) and FAS2240-2 (8.1.4-7m) at each site. The 2240-2 systems have a FlashPool accelerated aggregate and the customer is happy with the performance.
The new filers will be either a FAS2552 with additional shelves or we may ditch the entry toasters and go with a FAS8020 system and shelves at each site. If we go with the 2552 we’ll keep using FlashPool. If we go with the 8020 we have the option of going FlashCache.
We’d like to evaluate the potential performance improvements of using FlashCache on the 8020 in order to help with the purchasing decision. However, reading TR-3801 it looks like the FAS2000 and 2200 platforms don’t support Predictive Cache Statistics (PCS). This is confirmed by trying to enable PCS:
toaster> options flexscale.enable on FlexScale PCS is not active and cannot be enabled.
Are there any other tools which can help us estimate the performance benefits of FlashCache? Or perhaps some key metrics we can collect and analyse?
Cheers Steve ________________________________ This e-mail message has been scanned for Viruses and Content and cleared by MailMarshal ________________________________
DISCLAIMER: This email communication does not create or vary any contractual relationship between LogicalisCI and you. Internet communications are not secure and accordingly LogicalisCI does not accept any legal liability for the contents of this message. The contents of this email are confidential to the intended recipient at the email address to which it has been addressed. It may not be disclosed to or used by anyone other than this addressee, nor may it be copied in any way. If received in error, please contact LogicalisCI on the above number (listed as Main) quoting the name of the sender and the addressee and then delete it from your system. Please note that neither LogicalisCI nor the sender accepts any responsibility for viruses and it is your responsibility to scan the email and attachments (if any). Please be aware that LogicalisCI may monitor email traffic data and also email content for security purposes.
The following companies are collectively referred to in the above statements as LogicalisCI:
Logicalis Channel Islands Limited, Registered in Jersey No: 67161, Registered Office: Rue A La Dame, Five Oaks, St. Saviour, Jersey, JE2 7NH
Logicalis Jersey Limited, Registered in Jersey No: 10124, Registered Office: Rue A La Dame, Five Oaks, St. Saviour, Jersey, JE2 7NH
Logicalis Guernsey Limited, Registered in Guernsey No: 10896, Registered Office: Pitronnerie Road, St. Peter Port, Guernsey, GY1 2RF
________________________________ _______________________________________________ Toasters mailing list Toasters@teaparty.netmailto:Toasters@teaparty.net http://www.teaparty.net/mailman/listinfo/toasters
-
Borzenkov, Andrei -
Michael Bergman -
Stephen Stocke -
Tony Bar