Re: avoiding file system caching of a table

On Mon, Feb 17, 2014 at 5:09 PM, "Gabriel E. Sánchez Martínez" <gabrielesanchez@gmail.com> wrote:

On 02/16/2014 10:33 PM, Gabriel Sánchez Martínez wrote:

Is there a way of asking PostgreSQL to read the files of a table directly off the disk, asking the OS not to use the file cache?  I am running PostgreSQL 9.1 on Ubuntu Server 64-bit.  The server in question has the maximum amount of RAM it supports, but the database has grown much larger.  Most of the time it doesn't matter, because only specific tables or parts of indexed tables are queried, and all of that fits in the file cache.  But we have a new requirement of queries to a table several times larger than the total RAM, and the database has slowed down considerably for the other queries.

I am assuming that with every query to the large table, the OS caches the files containing the table's data, and since the table is larger than total RAM, all the old caches are cleared.  The caches that were useful for other smaller tables are lost, and the new caches of the large table are useless because on the next query caching will start again from the first files of the table. Please point out if there is a problem with this assumption.  Note that I am refering to OS file caching and not PostgreSQL caching.

If you told postgresql to tell the kernel not to cache the data it reads, how would this help?  The data you want in cache would no longer be pushed out of the cache, but that is because it would no longer be there in the first place.  You would have to make this instruction to the kernel be selective.  It would only tell it not to cache when it is doing a very large query.  It might be theoretically possible to do this, but it it would probably cause more harm than good to most people most of the time.

I read that the suggestion not to cache a file when reading it is given by programs at the time the file is opened.  That prompted me to think that there might be a way of telling PostgreSQL to apply that to the pages of a specific relation.  I did not mean to suggest it should be a process-wide or database-wide setting. 

 

Is there a way around this?  I have read that there is a way of asking the OS not to cache a file when the file is opened.  Is there a way of telling PostgreSQL to use this option when reading files that belong a specific table?

What about putting the table on a tablespace that is on a different device partition with the sync mount option?  Would that help?

I have read forum postings saying that the sync option affects writes, and will not prevent reads from caching.  At some forum posting I came across nocache, a utility for linux.  It is used by typing "nocache <command>" in a shell.  But I can't do that with a postgres process when a connection opens because postgres is the one opening the process.

You would have to start the entire service with that utility, then.  Which again would defeat the purpose.

Since a process is launched every time a session opens, e.g. a query window in pgAdmin, I thought it would be possible to do it per session rather than for the whole service.  Either way, I agree this wouldn't solve the problem.

 

 Does someone know a work-around, or a different solution to the problem?  Shouldn't PostgreSQL be smart about this and based on the statistics collected for a table and on the query plan know the harm that will be done if all of a very large table's pages are read and flush the cache?

PostgreSQL does know this.  It has a special ring "buffer access strategy" that it uses to prevent a large sequential scan from pushing all of the other data out of its shared_buffers.   It sounds like it is the kernel which is failing to employ similar logic on the file cache which the *kernel* manages.

I have no idea how the kernel manages its cache, but I think that since individual pages that store the data of the table are small, the kernel has no way of knowing that a process will read a very large number of relatively small files that collectively will cause harm.  Maybe if it were a single file large than total physical RAM it would act differently.  But I am just speculating. 

The kernel does also have some logic to prevent this, but it may or may not be very effective in your case (you haven't us what version of the kernel you are using). 

Thanks for asking.  3.8.0-35-generic.  I'm curious.  What does it do?  Or do you know where I can read about this (just out of curiosity). 

In fact one effort of the kernel to fix this problem for cases like yours ended up making it worse for other conditions, i.e. when the file being read sequentially was less than available RAM but greater than 1/2 available RAM.

You could try increasing shared_buffers (you haven't told us what it is set to now) until it takes up a big chunk of RAM, so that PostgreSQL manages more of the cache and the kernel manages less of it.  Setting it like that has been reported to cause problems on write-heavy work loads, but I haven't heard of problems on read-mostly workloads.

This server is read-mostly.  It has 64 GB of RAM, a single 6-core i7 processor, and four SATA hard drives on software RAID 10.  I get about 400 MB/s of sequential reads on simple benchmarks.  Shared buffers is set to 16 GB, temp buffers to 8 MB, work mem to 100 MB, and maintenance work mem to 100 MB.  I could probably tweak this but I know very little about it.  Do you think I should set any of these higher?  Other things run on the server, but most of the usage is PostgreSQL queries on tables of several hundred GB... some of which need to process whole tables. 

So other than the possibility of tweaking shared_buffers, the only other solution is getting a server with TBs of RAM?

Cheers,

Jeff

I really appreciate your help.  Thank you (and Tom).