Ying Zhang wrote:
Hi Maurice,
Stupid me. I only described how the current XRPC implementation is, but forgot to mention the ongoing work.
On Mar 28, 2008, at 09:35 , Keulen, M. van (Maurice) wrote:
Hi Peter,
Does it also mean that if I send a stream of messages to one server that all carry the same request-ID without waiting for a result of a message before sending out the next, that all are handled in the correct order and each is handled after completion of the previous one? In that case, it meets all my requirements. In fact, it gives much more than I ask for :-) I definitely like to follow any developments for this one!
This can be easily supported by adding a sequence number to each request message, I think.
But, if you have an updating request, and then a read-only request, do you then want to make the updates made by the first request visible to the second request?
Comparing it to the behaviour of a relational DBMS doing 'begin transaction ; update ; select ; commit' then, yes, I expect the effect of the update to be visible for the select, but not visible to select's done in other concurrently running transactions (the I of isolation in ACID). Maurice. -- ---------------------------------------------------------------------- Dr.Ir. M. van Keulen - Assistant Professor, Data Management Technology Univ. of Twente, Dept of EEMCS, POBox 217, 7500 AE Enschede, Netherlands Email: m.vankeulen@utwente.nl, Phone: +31 534893688, Fax: +31 534892927 Room: ZI 3039, WWW: http://www.cs.utwente.nl/~keulen