<div dir="ltr">Hey guys, thank you for taking the time to watch my talk. <br><div class="gmail_quote"><div dir="ltr"><br><div class="gmail_extra"><div class="gmail_quote"><div class="im"><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204);border-left-style:solid;padding-left:1ex">
<br>
Thanks for sharing the recording. Finally I've found out what's the<br>
purpose of total order protocol, although the drawbacks with WSC look<br>
pretty limiting. I'd really like to see the impact with WSC on, but that<br>
would require Infinispan itself as the main problem TOA is trying to<br>
solve is degraded performance due to deadlocking.<br></blockquote><div><br></div></div><div>The original paper [1] that describes the TOA algorithm, and its integration with Infinispan, has extensive performance figures comparing 2PC with Total order based commit. The results of their experiments show, that with the write/skew check in place, throughput increases and latency decreases when using the Total Order based commit in clusters containing six or more nodes. The transaction abort rate is similar to that of 2PC (when write/skew check is used). </div>
<div class="im">
<div><br></div><div> <br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204);border-left-style:solid;padding-left:1ex">
My main questions is about the ordering box size - if I understand that<br>
correctly, the ordering box consists of several physical machines.Then,<br>
what's the way to get the synchronized GSN from them? Is this the same<br>
as with regular TOA, taking the highest timestamp from all of the nodes?<br>
Or is there any internal synchronization within the ordering box?<br></blockquote><div><br></div></div><div>The synchronised GSN is maintained using internal synchronisation between the box members. Internal synchronisation is used, because we want the order box approach to be independent of the underlying total order protocol. Below is a brief description of how the GSN is maintained. <br>
<br>All box members start with a GSN value of 0. Upon receiving an ordering request from an Infinispan node, a box member sends a totally ordered message (containing the id of the infinispan message that requires ordering) to all box members (including itself). Because this box message is totally ordered, it is guaranteed that all box members receive the box message in the same order, therefore each box member assigns the same GSN to the infinispan message associated with this box message. As soon as the contacted box member has delivered its own message it will know the GSN of the infinispan message, and therefore a response (containing the GSN) can be sent to the infinispan node that requested ordering. <br>
<br>My intention is to prepare a document explaining in detail how our system works, which I will distribute to the mailing list at some point next week. </div><div><br></div><div>Ryan</div><div><br></div><div>[1] <a href="http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=6133071" target="_blank">http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=6133071</a></div>
<div><div class="h5">
<div> </div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204);border-left-style:solid;padding-left:1ex">
Anyway, I like the idea of centralizing things, as the problem itself is<br>
centralized. And as my parallel systems teacher said, solving<br>
centralized problem in distributed fashion usually causes more trouble<br>
than it's worth.<br>
<br>
Radim<br>
<br>
On 11/15/2013 12:29 PM, Paul Robinson wrote:<br>
> All,<br>
><br>
> Here's the recording from the Newcastle JBug...<br>
><br>
> <a href="http://www.youtube.com/watch?v=04qNcovQKLA" target="_blank">http://www.youtube.com/watch?v=04qNcovQKLA</a><br>
><br>
> I've also added Ryan to the CC, in case you have any specific comments<br>
> you want to direct his way.<br>
><br>
> Paul.<br>
><br>
> --<br>
> Paul Robinson<br>
> Web Service Transactions Lead<br>
> <a href="mailto:paul.robinson@redhat.com" target="_blank">paul.robinson@redhat.com</a> <mailto:<a href="mailto:paul.robinson@redhat.com" target="_blank">paul.robinson@redhat.com</a>><br>
><br>
> JBoss, a Division of Red Hat<br>
> Registered in England and Wales under Company Registration No. 03798903<br>
> Directors:Michael Cunningham (US), Charles Peters (US), Matt Parson<br>
> (US), Paul Hickey (Ireland)<br>
><br>
><br>
><br>
> _______________________________________________<br>
> infinispan-dev mailing list<br>
> <a href="mailto:infinispan-dev@lists.jboss.org" target="_blank">infinispan-dev@lists.jboss.org</a><br>
> <a href="https://lists.jboss.org/mailman/listinfo/infinispan-dev" target="_blank">https://lists.jboss.org/mailman/listinfo/infinispan-dev</a><br>
<br>
<br>
--<br>
Radim Vansa <<a href="mailto:rvansa@redhat.com" target="_blank">rvansa@redhat.com</a>><br>
JBoss DataGrid QA<br></blockquote></div></div></div></div></div>
</div><br></div>