Hi All!
I'm interested in running memcached on top of a block device (high
perf solid state PCIe thingy) to increase cache capacity but still
maintain decent latencies. I don't care about the non-volatility of
this device as much, just the size and speed. Has any work been do
On Sep 11, 2:57 pm, Mitch wrote:
> this device as much, just the size and speed. Has any work been done
> towards pointing memcached's cache at a block dev? I poked around at
> the source briefly.. looks like it plays much better with RAM (as
> designed).. but maybe mmap wouldn't be out of th
I think we're playing with the same "thingie" :)
Where can I find this particular branch?
On Fri, Sep 11, 2009 at 3:03 PM, Dustin wrote:
>
>
> On Sep 11, 2:57 pm, Mitch wrote:
>
>> this device as much, just the size and speed. Has any work been done
>> towards pointing memcached's cache at a b
On Sep 11, 3:24 pm, Dan Simoes wrote:
> I think we're playing with the same "thingie" :)
> Where can I find this particular branch?
I think it's a little behind and needs some merging to catch up.
Trond can speak for what the most current authoritative version is.
I've got a todo list entry t
Hello I am sorry to bother you all. We would love to get a build or
command line options to build the SSD capable branch/fork for a Sparc
Solaris machine. Has it been tested on such a machine and does anyone
have the command line options?
Thanks!
On Sep 11, 3:29 pm, Dustin wrote:
> On Sep 11,
quick recipe for we folks
that are looking to play around with something now..
As well as provide code review/feedback for the patch that would apply to trunk.
Thanks!
Mitch.
-Original Message-
From: RogueWarrior
Date: Wed, 16 Sep 2009 10:05:13
To: memcached
Subject: Re: block device
He
ig chunk with malloc */
mem_base = malloc(mem_limit);
changes to
if (prealloc) {
/* Map everything to my block device special file
in a big chunk with mmap..
fd is a global file descriptor (to be closed on cleanup)
filename is the block devic
cate everything in a big chunk with malloc */
mem_base = malloc(mem_limit);
changes to
if (prealloc) {
/* Map everything to my block device special file
in a big chunk with mmap..
fd is a global file descriptor (to be closed on cleanup)
filename
> By doing so you will end up talking to your block device _every_ time you
> try to look up an item in the cache. You might want to consider keeping the
> data structures in memory instead...
well, mmap'ed pages are buffered by linux using the page cache, it is
possible the cac
Mitch Crane wrote:
By doing so you will end up talking to your block device _every_ time you
try to look up an item in the cache. You might want to consider keeping the
data structures in memory instead...
well, mmap'ed pages are buffered by linux using the page cache, it is
possibl
io's. that's a little bit worse than
I was hoping. So a more optimized version would keep the hash map and
linked list in core, and the cache entries itself on our block device?
Is there a logical place to split those allocations?
thanks!
Mitch.
o you will end up by replacing the page.
So a more optimized version would keep the hash map and
linked list in core, and the cache entries itself on our block device?
Is there a logical place to split those allocations?
In my prototype I had the item meta-data in memory, and stored the
thanks a lot for your help, Trond! that clears it up :)
I'm curious as to what results people have had with the block device driver
that is in the community branch (or so I've been told). We're looking at
the Violin Memory device, primarily for DB storage, but it may have
interesting applications for memcache as well.
Thanks.
Dan
CPU wise maintaining the LRUs, etc)
So uh, yeah I don't get your motivation.
On Sat, Oct 25, 2008 at 8:08 PM, Dan Simoes <[EMAIL PROTECTED]> wrote:
> I'm curious as to what results people have had with the block device driver
> that is in the community branch (or so I&
Violin. The
Violin people say that when you factor in the operational costs (space,
power, etc) their solution saves money.
Why do you think I would lose my disk blocks randomly on a fast block
device? According to their specs, it's got a RAID5-like design so you can
lose memory boards wit
This mailing list's about memcached, not using memory or using caches in
general.
You do not want to use memcached for what you're talking about it. If you
want to shard a block device over a bunch of machine's memory, LVM and NBD
and ram block devices are your friend. But that
ROTECTED]> wrote:
> Replying only to you so I don't clutter the list.
>
> Did you read my original inquiry? No one said I wanted to shard a block
> device.
> I specifically asked about memcache on a block device, and specifically the
> Violin.
>
> We're alrea
<mailto:[EMAIL PROTECTED]>> wrote:
Replying only to you so I don't clutter the list.
Did you read my original inquiry? No one said I wanted to shard a
block device.
I specifically asked about memcache on a block device, and
specifically the Violin.
We
19 matches
Mail list logo