Hi
is there a memory limit for the memcached process ? i want to run it
with -m 4096 option , is that possible ?
actually ..i wanted to procure more ram. but someone just said that it
doesn't allow more than 2 gig. thats why i wanted to confirm. right now i am
on a 2 gig machine. but want to run memcached with 4 gigs of ram
thanks
On Fri, Mar 20, 2009 at 12:31 PM, Dustin dsalli...@gmail.com wrote:
On
it has 8 gigs of ram and its 64 bit
its going to be dedicated to memcached
thanks
sudipta
On Fri, Mar 20, 2009 at 12:38 PM, Clint Webb webb.cl...@gmail.com wrote:
Well if the machine had 4gb of ram, I wouldn't use 4096 for the -m option.
As I am sure that the OS and other bits are probably
On Mar 20, 2009, at 8:10 AM, Sudipta Banerjee wrote:
it has 8 gigs of ram and its 64 bit
Compile memcached as a 64bit binary and you should be able to use as
much memory as you want (I have tested with up to 30Gb)... Please note
that memcached use more memory than the memory specified
By the way, has anyone already tried to quantify in a *more
scientific* manner the actual extra memory we should let on a box to
be sure memcached won't swap ?
I guess memcached related overhead should be proportional to the
expected number of connection and their traffic. I guess this is about
any better chance for the 1.3 if I make it as a possible 0 extra cost
feature like the CAS? ;-)
Jean-Charles
On Mar 20, 8:20 am, dormando dorma...@rydia.net wrote:
We're going to pass on the patch for now. It's a bit much of a corner case
for the general release.
thanks to folks for taking
JC wrote:
any better chance for the 1.3 if I make it as a possible 0 extra cost
feature like the CAS? ;-)
I would say that it is too late for inclusion in the initial release of
1.3. I would say that we should only apply bugfixes to the 1.3 right now
and get it into a stable state so
I'd just like to check that my understanding of how a distributed
memcached system works. Are these 2 assertions correct?
1) It's the client which is configured to know about multiple
memcached servers; the servers themselves know nothing about being
part of a cluster.
2) If you have one client
Assertion 1 is correct.
Assertion 2 and the answer to your two questions depends on the client and
the type of hashing they use to determine which server to use. To understand
this, check out the libketama consistent hash
3. There is not a single hashing algorithm defined for use by a
memcached client.
so
4. There is no guarantee that two clients with the same configured set
of servers, configured in the same order, will use the same server for
the same key.
If they use the same client software, they will
Hi Rachel,
For assertation 2, yes, it's a matter of chance which key is stored in which
server, but depending on which server selection algorithm you use, you can
determine the probability of a key being stored on the same server from both
clients. (It's higher if you use Ketama) Still, you
Thanks for your replies. So is this correct?
3. There is not a single hashing algorithm defined for use by a
memcached client.
so
4. There is no guarantee that two clients with the same configured set
of servers, configured in the same order, will use the same server for
the same key.
Rachel
It's at least a couple megs. It depends on how many parallel connections
you have, and the typical size of your read buffers.
If you do a lot of large multi-gets, you'll use more ram than otherwise.
Future releases will probably track this memory more closely? It's not
hard to do.
-Dormando
On
On 3/20/09 4:17 PM, meppum wrote:
Recently, I started to look closer at how our cluster is performing. I
have noticed that it does not look like Memcached is as naturally
balanced as it should be.
It is not balanced. It is psuedo-random.
We run ten Memcached instances across ten physical
On Mar 20, 5:59 pm, dormando dorma...@rydia.net wrote:
iirc we were looking at using an option like that to pass startup
parameters into the storage engine interface. would that be the -o option,
then memcached itself takes normal parameters? Or would we have two
separate long format
On Mar 20, 5:30 pm, dormando dorma...@rydia.net wrote:
Think it's worth me floating the compatability change by
planetmysql/twitter/etc to see if we can dig up more people with shitty
apps?
IIRC the only real functionality that changed was... you could kind of
store a word on the front or
Words by meppum [Fri, Mar 20, 2009 at 07:02:08PM -0700]:
Brian,
Thanks for the reply. My responses are below.
On Mar 20, 5:43 pm, Brian Moon br...@moonspot.net wrote:
On 3/20/09 4:17 PM, meppum wrote:
Recently, I started to look closer at how our cluster is performing. I
have
On 3/20/09 9:33 PM, meppum wrote:
More requests, more memory, and more keys (i believe this is
curr_items in stats). I am seeing a difference of 7-9% in the number
of keys between the server with the most and the server with the
least.
I'm sure there are hot keys, thought I haven't really
18 matches
Mail list logo
