Hi Aniruddha, Are you using trunk for your tests?

On Sep 21, 2012, at 4:42 AM, Aniruddha Laud wrote:

> I'm logging latencies in the per channel bookie client by overriding the
> read completion and add completion callbacks.
> 
> When a hedwig subscriber subscribes to a topic and it hasn't consumed any
> messages for a long time, it has to issue requests to bookkeeper to read
> entries. We are observing very high latencies during these operations.
> 
> throttle limit for the bookkeeper client is set at 5000. We are using the
> Hierarchical ledger manager.
> 

We throttle before PerChannelBookieClient, so if the high latency you're 
observing is in PCBC, then it shouldn't be related to throttling.

> The bookies have 5 disks. 1 for the journal, 4 for the ledgers. On the
> bookie server pageSize=32768, open file limit = 20000. I checked a few
> bookies and the number of open ledgers was around 5000 on each. However,
> the add and read latencies on the servers were less than 1 second (measured
> using jconsole and the exposed jmx stats).
> 

Did you mean to say 1ms? 1 second for a read/write is a lot.

Can you give some more detail about your traffic, like the mix of reads and 
adds, size of entries, etc.?

> Also, the bookie disks are standard spin drives and were doing about 40MBps
> reads and 10MBps writes when measured with dstat. Any thoughts would be
> helpful.

Seems low to me, although I would think that they are not the culprits. 

-Flavio

Reply via email to