There are archived discussions floating around about this subject,
particularly with SSD. Are there open source works (implementation,
src code, interface description, etc) available for people to take a
look ?
Thanks,
Wendy
On Mon, Jan 9, 2012 at 1:21 PM, Wendy Cheng wrote:
> There are archived discussions floating around about this subject,
> particularly with SSD. Are there open source works (implementation,
> src code, interface description, etc) available for people to take a
> look ?
>
Found an
On Mon, Jan 9, 2012 at 2:21 PM, dormando wrote:
>> There are archived discussions floating around about this subject,
>> particularly with SSD. Are there open source works (implementation,
>> src code, interface description, etc) available for people to take a
>> look ?
>
> There is a working and
On Mon, Jan 9, 2012 at 2:28 PM, Dustin wrote:
>
> On Monday, January 9, 2012 2:13:36 PM UTC-8, Wendy Cheng wrote:
>>
>> Found an expired RFC at
>> http://code.google.com/p/memcached/wiki/EngineInterface. Look pretty
>> old though. Wondering whether there are (open s
On Tue, Jan 10, 2012 at 12:10 AM, Dustin wrote:
>> By taking a quick browse on the source, it seems only "default_engine"
>> is included (?) that uses RAM as a backend. Is this right ? So I
>> assume for other type of backends, we have to supply our own engine by
>> implementing the "ENGINE_HANDLE
On Tue, Jan 10, 2012 at 7:42 PM, Dustin wrote:
>
> On Tuesday, January 10, 2012 4:19:26 PM UTC-8, Wendy Cheng wrote:
>
>>
>> It seems the engine operations (hooks) are invoked as part of the
>>
>> network request processing logic (e.g. process_bin_set() et
On Wed, Jan 11, 2012 at 10:51 AM, Dustin wrote:
> One thing that might not be obvious is that you don't have to use an item
> structure to store your data. You just need it to transport it between the
> back and front.
First, I need to pass a thank-you note for all the helps received, as
well
Does anyone have a ballpark number for "memslap --server
--test=set" command ? Somehow I got ~0.6 seconds that is surprising.
Intuitively, our network round-trip latency for 16K tcp data is less
than 0.001 second. I expected the memslap "set" command would not go
up to msec range (?). Our lab mach
On Tue, Feb 28, 2012 at 9:47 AM, Wendy Cheng wrote:
> Does anyone have a ballpark number for "memslap --server
> --test=set" command ? Somehow I got ~0.6 seconds that is surprising.
> Intuitively, our network round-trip latency for 16K tcp data is less
> than 0.001 second.
On Tue, Feb 28, 2012 at 10:02 AM, Wendy Cheng wrote:
> On Tue, Feb 28, 2012 at 9:47 AM, Wendy Cheng wrote:
>> Does anyone have a ballpark number for "memslap --server
>> --test=set" command ? Somehow I got ~0.6 seconds that is surprising.
>> Intuitively, our net
On Tue, Feb 28, 2012 at 11:01 AM, dormando wrote:
>
> You might want to consider mc-crusher as well.
> https://github.com/dormando/mc-crusher
>
> it's still rough, but from what it can do it's very flexible.
Thanks ! Will take a look soon ... Wendy
Hi,
We plan to start tuning work for our Memcached setup. The binaries
were built on top of "engine_pu" branch (storage engine interface). A
quick test run (everything on default, including storage engine
itself) using
"/usr/local/bin/memslap --server xyz --test=get --concurrency=1024"
got a bun
On Thu, Mar 1, 2012 at 5:55 PM, Wendy Cheng wrote:
> Hi,
>
> We plan to start tuning work for our Memcached setup. The binaries
> were built on top of "engine_pu" branch (storage engine interface). A
> quick test run (everything on default, including storage engine
> i
On Fri, Mar 2, 2012 at 2:51 PM, Ripduman Sohan wrote:
> On 2 March 2012 01:55, Wendy Cheng wrote:
>> Hi,
>>
>> We plan to start tuning work for our Memcached setup. The binaries
>> were built on top of "engine_pu" branch (storage engine interface). A
>
On Tue, Feb 28, 2012 at 3:51 AM, Slawomir Pryczek wrote:
> Actually i have already modified it and it seems working. It's
> compatible with current client libraries. I'll post on mailing list
> with some simple benchmarks if anyone's interested.
>
I would be interested to learn the numbers if po
As part of the tuning work, it is noticed that the maximum connection
count is controlled by "settings.maxconns". There is a comment next to
the default setting:
settings.maxconns = 1024; /* to limit connections-related
memory to about 5MB */
I'm wondering how the 5MB was calculated. Cou
-- Wendy
>
> --
> Blog: http://www.chenzongzhi.info
> Twitter: https://twitter.com/baotiao
> Git: https://github.com/baotiao
>
> On 2012-3-8, at 上午8:40, Wendy Cheng wrote:
>
> As part of the tuning work, it is noticed that the maximum connection
> count is controlled by
This mutex seems to only exist with Storage Engine branch.
We're working on prototyping a new storage engine on top of an NVRAM
device. This morning, the test server hangs. Four worker threads are
blocked waiting for cache_lock. The cache_lock holder is blocked in
"notify_io_complete()" waiting fo
On Wed, Mar 14, 2012 at 11:02 AM, Wendy Cheng wrote:
> This mutex seems to only exist with Storage Engine branch.
>
> We're working on prototyping a new storage engine on top of an NVRAM
> device. This morning, the test server hangs. Four worker threads are
> blocked waiting
On Sun, Mar 25, 2012 at 12:16 PM, dormando wrote:
>> Thanks. Any reason to move item_lock(s) back to cache_lock ?
>>
>> We've done preliminary performance runs using memaslap. Memory limits
>> are altered from 32MB up to 2GB. The total transaction-per-second
>> numbers drop as the memory limits ar
On Wed, Apr 4, 2012 at 10:47 PM, Dustin wrote:
>
> On Wednesday, April 4, 2012 11:48:26 AM UTC-7, Aaron Stone wrote:
>>
>> Hello list! Long time. The storage engine branch looks a little bit behind
>> master - is github.com/memcached/memcached:engine-pu the most up to date
>> branch (last commit i
On Thu, Apr 5, 2012 at 10:39 PM, Dustin wrote:
>
> On Thursday, April 5, 2012 9:51:33 AM UTC-7, Wendy Cheng wrote:
>
>>
>> We also have a prototype storage engine (hybrid of RAM + flash memory)
>> up and running. Would be happy to help out
>
> That would be
On Wed, Apr 25, 2012 at 8:32 AM, pcdinh wrote:
> Hi all,
>
> I compiled and installed Memcached 1.4.3 from source:
> http://memcached.googlecode.com/files/memcached-1.4.13.tar.gz
>
> The configuration is quite simple: ./configure --prefix=/usr/local/
> memcached
>
> It works if I install it on a d
On Wed, Apr 25, 2012 at 9:25 AM, pcdinh wrote:
> Here is the list of libevent-related packages that are installed in my
> server
>
> root@spica:/usr/src/libevent-2.0.18-stable# dpkg --get-selections |
> grep libevent
> libevent-2.0-5 install
> libevent-core-2.0-5
On Thu, Apr 26, 2012 at 2:06 AM, pcdinh wrote:
> The native box is my development environment at my office. Sample OS
> (Ubuntu 11.10), same libs (autoconf 2.68 )
> In development environment it works. Now I want to move the source
> code to a production environment which is Xen-based
> I have
25 matches
Mail list logo