- 01 9月, 2011 1 次提交
-
-
由 bessani@gmail.com 提交于
Removed and refactored a lot of code, added new microbenchmarks and improved the parallelism between the delivery thread and server communication thread. Additionally, processing of out of context message changed a bit (done in the server communication thread, instead of the delivery thread). IMPORTANT: Service replica API changed!!! NOTE: currently achieving a zero-byte throughput of 151Kops/sec ;-)
-
- 23 8月, 2011 1 次提交
-
-
由 bessani@gmail.com 提交于
- Improvements on the performance of communication system - big clean up of the code (many old methods deleted) - corrected bug in out of context (now, a consensus is only started if a propose is received) Note: still need to verify if state transfer and leader change works right
-
- 25 6月, 2011 1 次提交
-
-
由 liquidsnake@sapo.pt 提交于
In this commit, I have submitted the code for the leader change. Unfortunately there is some kind of conflict between the code I developed for an older version of smart, and the most recent code available on the svn. Such conflict renders the leader change protocol faulty. Further debugging is necessary. The normal operating remains good, though.
-
- 10 3月, 2011 1 次提交
-
-
由 ChaosguardX 提交于
Fixed a bug in ExecutionManager.java where not all stoppedMsgs would be removed from the List upon restart
-
- 03 2月, 2011 1 次提交
-
-
由 liquidsnake@sapo.pt 提交于
In this commit, it has been added a timeout for the state transfer protocol, and fixed an implementantion error regarding f + 1 state hashes consistent amoung them, and a complete state inconsistent with those ashes. It was also fixed a bug that would crash clients when a a reconfiguration was applied, and also fixed 2 bugs related to the checkpoints, which would make replicas crash. Finally, there are also some lines of code that have been slightly modified, but which shouls not affect the behavior of SMaRt.
-
- 19 11月, 2010 1 次提交
-
-
由 liquidsnake@sapo.pt 提交于
Removed out of context thread. The processing out of context messages is now made at the delivery thread and Server CS thread Removed paxos internal timers. From now on, the external timers from the total ordering layer shall trigger the leader change Added a few debugging messages for delivering messages Fixed a bug related to signatures of forwarded messages from replicas to the leader Fixed a bug on the TTP that would make it crash in some cases Fixed 2 bugs on the state transfer protocol responsable for not letting the old leader fetch the state once it recovered
-
- 08 10月, 2010 1 次提交
-
-
由 liquidsnake@sapo.pt 提交于
In this commit, it was added a new feature to SMaRt. It now supports dynamic views. Some bugs from the previous commit were also fixed. This was a fairly extensive update to SMaRt.
-
- 01 10月, 2010 1 次提交
-
-
由 liquidsnake@sapo.pt 提交于
-
- 08 5月, 2010 1 次提交
-
-
由 ChaosguardX 提交于
Nonces are now generated for each TOMMessage in a Batch with a single seed that is sent along with the batch instead of a fixed number of nonces per Batch. Decided notification in Consensus.java now is done via a sync monitor instead of busy waiting on the decided value to be set. Because of the Java MemoryModel the former mechanism was problematic and could theoretically lead to the DeliveryThread never learning the new value - which is very unlikely though.
-
- 11 3月, 2010 1 次提交
-
-
由 liquidsnake@sapo.pt 提交于
In this commit I made an optimization to the state transfer protocol that consists on the sending of hash's of the state. Only one replica will send the actual state. I also worked towards fixing some bugs.
-
- 04 3月, 2010 1 次提交
-
-
由 liquidsnake@sapo.pt 提交于
In this commit I started to clean up the code that I wrote. I also added some debug messages which are produced with the Logger object.
-
- 27 2月, 2010 1 次提交
-
-
由 liquidsnake@sapo.pt 提交于
In this commit, I continued to work on the state transfer protocol .I finally managed to correctly implement the optimization for the revived replica to request the state sooner than the paxos highmark.
-
- 25 2月, 2010 1 次提交
-
-
由 liquidsnake@sapo.pt 提交于
In this commit, I continued to work on the state transfer protocol.I finally tested the protocol in the case of a crash of a replica. I was able to make the replica recover the state on those circunstances. However, the replica will still only requests the state after it passes the highmark value. I started working on an optimization to request the state before it reaches that highmark, but its still not completed.
-
- 23 2月, 2010 1 次提交
-
-
由 liquidsnake@sapo.pt 提交于
In this commit, I continued to work on the state transfer protocol. It is already capable of transfering leader information, and supply it to the delayed replica. This means the code is no longer hacked to always assume replica 0 is the leader. It also does garbage collection again. I also made some minor adjustements to some debbuging messages and the new demo
-
- 13 2月, 2010 1 次提交
-
-
由 liquidsnake@sapo.pt 提交于
In this commit, i've continued to try to find and fix more errors related to the updating of delayed replicas. I have managed to fix three of them, but the code is merely "patched". If there is a leader change, SMaRt will go haywire. This version o the code should not be used for other purpose then testing. I added more debug messages, and I will keep looking at the code during the weekend. I expect to fix some more bugs that i found during that period. I will also produce more debug logs so i can try to find more bugs.
-
- 09 2月, 2010 1 次提交
-
-
由 liquidsnake@sapo.pt 提交于
In this commit, i've continued to try to find the errors on the state transfer protocol. I think i found one, and I'm beginning to add code to fix it. However, I would like to speak to the teachers tomorrow about this. P.S. I also added more debug messages
-
- 08 2月, 2010 1 次提交
-
-
由 liquidsnake@sapo.pt 提交于
I this commit, I continued the work on the state transfer protocol. I implemented the modifications discussed. However, the protocol remains faulty. Further debugging is required.
-
- 06 2月, 2010 1 次提交
-
-
由 liquidsnake@sapo.pt 提交于
In this commit, i've only added some debug messages.But I committed it anyway, so I can work on SMaRt during the weekend at home
-
- 04 2月, 2010 1 次提交
-
-
由 liquidsnake@sapo.pt 提交于
In this commit, i've continued my task of debuging the code.I finally managed to get the replica to update itself. However, when the clients finish to send requests, that same replica still remains a little behind, even though it sucessfully updates itself during previous executions of the state transfer protocol.
-
- 29 1月, 2010 1 次提交
-
-
由 liquidsnake@sapo.pt 提交于
In this commit, i've continued my task of debuging the code. I believe the protocol itself is functioning correctly, but I still have to set the replicas to actually enforce the received state upon the application. The commited code is tested, and SMaRt seems to work properly according to the counter demo.
-
- 28 1月, 2010 1 次提交
-
-
由 liquidsnake@sapo.pt 提交于
In this commit, i've continued my task of debuging the code. I made sure that the checkpoints are indeed working properly, so i re-activated the code which implements it. Part of the state transfer protocol is also working, but I committed it de-activated. I'll resume the debugging tomorrow.
-
- 26 1月, 2010 1 次提交
-
-
由 liquidsnake@sapo.pt 提交于
I've resumed my work on the state transfer protocol. However, I'm having trouble getting it to work right. I've de-activated the protocol, in order to safely commit the changes.
-
- 19 1月, 2010 1 次提交
-
-
由 liquidsnake@sapo.pt 提交于
I have completed the coding of the state transfer protocol. However, I'm still yet to test the changes made in this commit. Therefore, this is likely to be an unstable version of SMaRt
-
- 08 1月, 2010 1 次提交
-
-
由 liquidsnake@sapo.pt 提交于
Previous changes to the message handler were not intended. It were really meant in the execution manager. Fixed.
-
- 07 1月, 2010 1 次提交
-
-
由 liquidsnake@sapo.pt 提交于
-
- 14 12月, 2009 1 次提交
-
-
由 pjsousa@gmail.com 提交于
-