Dear Jeff,

CORBA uses IIOP protocol which encodes the data into binary format.

I have referred few white papers in which CORBA is being replaced by zeromq
and protocol buffers, but it was twice slower in performance as compared
with CORBA.

I am sure i can use zeromq and protocol buffers but which pattern will be
best fit for my kind of problem.

shall i use multithreaded REQ and REP pattern or shall i use PUB/SUB or
 PUSH/PULL.?

Regards,
Yogesh



On Thu, May 26, 2016 at 7:13 PM, Jeffrey Getzin <[email protected]> wrote:

> Think of protocol buffers as serializing all your data into efficient
> binary data. In other words, it's not like JSON or XML, which are
> represented as text. Therefore, the protobufs will generally be much
> smaller than with other schemes. However, I don't know how CORBA encodes
> data, so I can't be sure protobufs will be more efficient. My GUESS is that
> they will, because Google is behind them, and they're used everywhere in
> Google's incredibly high-scaling applications.
>
> Then let's look at ZeroMQ. ZeroMQ lists itself as being asynchronous,
> which means that your app should not hang. Furthermore, I don't think
> you'll need to break up your messages because that's what TCP/IP is
> supposed to do under the hood. In other words, your ZeroMQ will send a
> message over TCP/IP/UDP/etc., and the underlying transport layer may or may
> not break those messages into smaller packets.
>
> The upshot is that while I don't have specific information about CORBA,
> your application, or even ZeroMQ, I'm GUESSING that yes, this application
> will be much more to your liking than your old approach. However, you
> really need to do a few experiments with real data to be sure. I would not
> commit to ANY new technology, even protobufs, until I was sure that they
> not only worked well, but worked well for my individual situation and needs.
>
>      Jeff
>
>
> On Thu, May 26, 2016 at 8:19 AM, Yogesh Fulsunge <
> [email protected]> wrote:
>
>> Dear Jeff,
>>
>> The Problem with CORBA is that it hangs while sending large amount of
>> data lets say 15 MB
>>
>> so this problem can be solved by serialisation using  using protocol
>> buffer and transport using zeromq??
>>
>> also do i need to encode15 MB blob in one request or in multiple request?
>>
>> can you please clarify wht will be approach for encoding?
>>
>> Regards,
>> Yogesh
>> 9960352599
>>
>>
>> On Friday, May 20, 2016 at 7:03:19 PM UTC+5:30, Jeffrey Getzin wrote:
>>>
>>> I don't see why not, but what type of data will you be transferring?
>>> Some data will transfer more efficiently than others. For example,I suspect
>>> that sending messages with a single large BLOB would not be as efficient as
>>> sending many text fields.
>>>
>>>       Jeff
>>>
>> --
>> You received this message because you are subscribed to a topic in the
>> Google Groups "Protocol Buffers" group.
>> To unsubscribe from this topic, visit
>> https://groups.google.com/d/topic/protobuf/VvcEZwVlGPM/unsubscribe.
>> To unsubscribe from this group and all its topics, send an email to
>> [email protected].
>> To post to this group, send email to [email protected].
>> Visit this group at https://groups.google.com/group/protobuf.
>> For more options, visit https://groups.google.com/d/optout.
>>
>
>

-- 
You received this message because you are subscribed to the Google Groups 
"Protocol Buffers" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To post to this group, send email to [email protected].
Visit this group at https://groups.google.com/group/protobuf.
For more options, visit https://groups.google.com/d/optout.

Reply via email to