Re: Parallel serialization and writes ? (Was: ACIDity of Prevayler 2.6)

On Sun, Nov 18, 2012 at 2:09 AM, Arnaud Masson <[email protected]> wrote:

> 1) 50ms-100ms seems a large value (especially on SSD).
>

Hmm. SSD does change the picture a bit. :) But 100ms is definitely what I
remember from running the Prevayler scalability tests on a hard disk 5+
years ago. I remember getting throughput maxing out close to 1000
transactions per second, with roughly 100 per batch and 10 batches per
second (i.e. 100ms each).

2)  "Adding ... latency to _every_ query"
> This is the worst case scenario, isn't it?
>

Ha, yeah. Though we could think of it as the _best_ case, because it means
your app is popular. :)

With a 100ms transaction write latency, it just takes 10 POSTs per second
to affect every GET. If an SSD allows, say, a 10ms transaction write
latency, then it would take 100 POSTs per second to affect every GET, which
I admit is a pretty intense load. I think typical webapps have roughly 1%
to 10% of requests being POSTs, so you're right that it might not be an
issue as long as you're not getting thousands of requests per second.

------------------------------------------------------------------------------
Monitor your physical, virtual and cloud infrastructure from a single
web console. Get in-depth insight into apps, servers, databases, vmware,
SAP, cloud infrastructure, etc. Download 30-day Free Trial.
Pricing starts from $795 for 25 servers or applications!
http://p.sf.net/sfu/zoho_dev2dev_nov

_______________________________________________
To unsubscribe go to the end of this page: http://lists.sourceforge.net/lists/listinfo/prevayler-discussion
_______________________________________________
"Databases in Memoriam" -- http://www.prevayler.org