Benchmarking binprot is nearly useless: there's nothing inherent in the
protocol which makes it any faster for raw requests/responses. It does
give your application more flexiblity on doing your gets/sets. IE: noreply
commands for issuing sets without waiting for the response. You can also
pack
While things like memslap and such will show very large gains, your
application may never see them. In my testing of my application, using
the binary protocol showed no noticable difference in performance. That
is simply because memcached is not the bottleneck in my application. I
highly doubt
Have there been any benchmarks about packet sizes for text vs binary
communication to memcached? And, what sort of general performance gains one
can expect by switching to binary protocol.
Thanks!
-Manish
--
---
You received this message because you are subscribed to the Google Groups
You can use memtier_benchmarkhttps://github.com/GarantiaData/memtier_benchmark
to
test the performance differences
On Thu, Jul 18, 2013 at 4:56 AM, Manish Jain man...@quora.com wrote:
Have there been any benchmarks about packet sizes for text vs binary
communication to memcached? And, what