Re: Intro

as children. It is like having a file-system (in each server) that allows a file to also be a directory. Regards, Ibrahim EL-sanosi From: Mahmood Naderan Sent: 29 April 2018 16:53:01 To: user@zookeeper.apache.org Subject: Intro Hi About the figure represented

RE: Write Ahead Log

and Marco Serafini. "Zab: High-performance broadcast for primary-backup systems." Dependable Systems & Networks (DSN), 2011 IEEE/IFIP 41st International Conference on. IEEE, 2011. Ibrahim -Original Message- From: tj5527 [mailto:tj5...@protonmail.com] Sent: 08 Jan

Re: Extremely different readings on different zookeeper deployments

This may help as well.. 1. The write request requres consensus (three communcation steps where read does not). 2. The more write requests the Zab has, the more messages (Forwards/Proposals/ACKS/Commits messages) will be generated/sent between Leader/Followers. 3. As (2) occurs, the replics (Leader

RE: november 17 zk meetup (additional registration info required)

Can we watch the event remotely? Sanosi -Original Message- From: Benjamin Reed [mailto:br...@apache.org] Sent: 16 November 2016 19:07 To: user@zookeeper.apache.org; DevZooKeeper Subject: november 17 zk meetup (additional registration info required) sorry for the late notice, but we ha

Re: Dealing with old delivered requests in memory

Thank you Patrick, Ibrahim On Wed, Mar 23, 2016 at 4:44 PM, Patrick Hunt wrote: > On Tue, Mar 22, 2016 at 7:53 AM, ibrahim El-sanosi > wrote: > > Thank you Patrick, > > > > > > > >>That's correct. ZK is an in-memory store. As such all znodes > &

Re: Dealing with old delivered requests in memory

es persent in memory at the moment). Does ZK servers reapply all snapshot in memory or just takes the latest one? Ibrahim On Tue, Mar 22, 2016 at 2:43 PM, Patrick Hunt wrote: > On Fri, Mar 18, 2016 at 10:10 AM, ibrahim El-sanosi > wrote: > > Ok, Thank you for answering the question.

Re: Dealing with old delivered requests in memory

: > Hi Ibrahim, > > Are you asking about how we compact old logs? We do it by taking snapshots > so that upon recovery, we only load the latest snapshot and replay the txn > log from the snapshot tag. The snapshot tag is the last zxid committed when > we start producing the snap

Re: Dealing with old delivered requests in memory

Thank you Jordan, So as "ZooKeeper stores the entire database in memory. ZNodes exist until deleted.". As we know the memory size is limited, we expect the OutOfMemoryError excption could occur even we set the Max heap size as maxumim value as possible? Ibrahim On Fri, Mar 18, 2016

Re: Dealing with old delivered requests in memory

. Thank you indeed Ibrahim On Fri, Mar 18, 2016 at 4:12 PM, Flavio Junqueira wrote: > Hi Ibrahim, > > Are you asking about how we compact old logs? We do it by taking snapshots > so that upon recovery, we only load the latest snapshot and replay the txn > log from the snapshot tag. T

Re: Dealing with old delivered requests in memory

size of ZK state reaches to the memory size). 2. If the leader crashes or lost a quorum, recovery mechanism applies latest snapshots. In this scenario, it is possible to eliminate the old ZK states and could save memory space. Is that right? Regards, Ibrahim On Fri, Mar 18, 2016 at 4:49 PM

Dealing with old delivered requests in memory

, what strategy ZK does to find and delete unused Znodes? Ibrahim

RE: 3-server Zab cluster

ieve better latency, we may provide this optimization (above algorithm) as optional. I hope my aim is clear now. Ibrahim -Original Message- From: Flavio Junqueira [mailto:f...@apache.org] Sent: Monday, October 05, 2015 06:23 م To: user@zookeeper.apache.org Subject: Re: 3-server Zab cl

RE: 3-server Zab cluster

e sure that such txns are recorded in a quorum before >establishing a new epoch, though. I guess you meant a quorum not a LOG in above world *log* !!! Thank you Ibrahim -Original Message- From: Flavio Junqueira [mailto:f...@apache.org] Sent: Monday, October 05, 2015 06:23 م To: user@zooke

RE: Zab Failure scenario

Thank you Flavio, it makes sense. Ibrahim -Original Message- From: Flavio P JUNQUEIRA [mailto:f...@apache.org] Sent: Sunday, October 04, 2015 05:43 م To: user@zookeeper.apache.org Subject: RE: Zab Failure scenario Acks aren't logged and neither are commits. A prospective leader co

RE: 3-server Zab cluster

http://zookeeper-user.578899.n2.nabble.com/Zab-Failure-scenario-td7581583.html for more info What do you think? Does anyone have any questions or concerns about such (small) optimization? Ibrahim From: Rakesh Radhakrishnan [mailto:rakeshr.apa...@gmail.com] Sent: Thursday, October 01, 2015 06

RE: Zab Failure scenario

d =10 did not get a quorum of ACKS from pervious epoch) Regards, Ibrahim -Original Message- From: Ibrahim El-sanosi (PGR) Sent: Sunday, October 04, 2015 03:48 م To: user@zookeeper.apache.org Subject: RE: Zab Failure scenario Does zxid = 10 commit because there are acknowledgments f

RE: Zab Failure scenario

, resulting in L (pervious leader) becomes a prospective leader, because of zxid =10. What happen then? Does zxid = 10 commit eventually before the end of synchronization phase or discard? Regards, Ibrahim -Original Message- From: Flavio P JUNQUEIRA [mailto:f...@apache.org] Sent: Sunday

RE: 3-server Zab cluster

Yes Alex, in my post I mentioned that this (small) optimization can only work with 3-servers cluster. Who could confirm the optimization can work? Ibrahim -Original Message- From: Alexander Shraer [mailto:shra...@gmail.com] Sent: Tuesday, September 29, 2015 12:11 ص To: user

RE: 3-server Zab cluster

Zxid = 10 committed. F1 forms a quorum with the L (pervious leader), F1 becomes new leader as it has latest zxid (10) . L (new leader) syncs its state with L (pervious leader now become a follower), as a result Zxid10 commits by new quorum. Is that correct? What do you think? Ibrahim ---

RE: Zab Failure scenario

, September 28, 2015 02:56 ص To: user@zookeeper.apache.org Subject: Re: Zab Failure scenario On 27 September 2015 at 10:12, Ibrahim El-sanosi (PGR) < i.s.el-san...@newcastle.ac.uk> wrote: > Thank you Flavio for explanation. It really makes sense for me. > > > I'm not sure why

3-server Zab cluster

in 3-servers cluster (follower reaches a majority as soon as it acks). Does anyone see any problems with such (small) optimization? Ibrahim

RE: Zab Failure scenario

e different solution for mentioned scenario. Ibrahim -Original Message- From: Flavio P JUNQUEIRA [mailto:f...@apache.org] Sent: Sunday, September 27, 2015 05:21 م To: user@zookeeper.apache.org Subject: Re: Zab Failure scenario In 3, it is not exactly a pending proposal, but if the leade

Zab Failure scenario

larger myid, So, 3. L sends a pending proposal with Zxid =10 to F1. 4. F1 logs and sends an ACK. 5. Upon receiving ACK, L commits p10 and sends an ACK. (1)Is this true or false? (2)Regards, Ibrahim

GM protocol

org.apache.zookeeper.server.quorum, do such job, is that correct? Regards, Ibrahim

RE: Doubt in Atomic Broadcast

. It is true that the leader election (looking phase) may be different but I am here comparing the protocols in failure-free period. Regards, Ibrahim -Original Message- From: Flavio Junqueira [mailto:fpjunque...@yahoo.com.INVALID] Sent: Sunday, August 30, 2015 07:47 م To: user

Latency performance in write-intensive workload.

Dear ZooKeeper team, I am looking for a graph shows ZooKeeper latency in write-intensive workload. Can you refer to any documents or white paper contains such information? Note that I can only find throughput performance in some documents, but no latency figures Thank you Ibrahim -- View

RE: Use cases of ZooKeeper

need is an eventually consistent kv store, why not use something >like Cassandra ? > >On Thu, May 21, 2015 at 10:41 AM, Flavio Junqueira < >fpjunque...@yahoo.com.invalid> wrote: > >> Hi Ibrahim, >> >> Consider having a look at the jzab kv example. >> >> -Fla

Use cases of ZooKeeper

it is possible to use ZooKeeper, is there any use cases that use ZooKeeper for such scenario? Ibrahim -- View this message in context: http://zookeeper-user.578899.n2.nabble.com/Use-cases-of-ZooKeeper-tp7581115.html Sent from the zookeeper-user mailing list archive at Nabble.com.

RE: Adding change to ZK

Hi Michi, How should I update the document, should I update it locally and send it to you all for feedback, or what is the best way to do that? Ibrahim >-Original Message- >From: Michi Mutsuzaki [mailto:mutsuz...@gmail.com] >Sent: 15 April 2015 07:06 >To: user@zookeepe

Re: Adding change to ZK

Yes Michi, I am happy to do that. Ibrahim Sent from my iPhone On 15 Apr 2015, at 07:06, Michi Mutsuzaki mailto:mutsuz...@gmail.com>> wrote: Agreed. Would you like to update the wiki page? On Tue, Apr 14, 2015 at 9:23 AM, Ibrahim mailto:i.s.el-san...@newcastle.ac.uk>> wrote: Than

Re: Adding change to ZK

the https://cwiki.apache.org/confluence/display/ZOOKEEPER/UsingEclipse) nice, simple, and clear to help users and (new) developers to obtain and play with the code. Thank you once again ibrahim <http://zookeeper-user.578899.n2.nabble.com/file/n7580971/packages2.jpg> <http://zookeeper-us

RE: Adding change to ZK

clover.info: clover: jute: BUILD FAILED D:\workspacezk\zookeeper6\tags\release-3.4.6\build.xml:272: Class not found: javac1.8 Total time: 780 milliseconds Any helps please Ibrahim From: Flavio Junqueira-2 [via zookeeper-user] [mailto:ml-node+s578899n7580962...@n2.nabble.com] Sent: 13 April 20

RE: Adding change to ZK

Flavio, The link doesn't work, do you mean this https://cwiki.apache.org/confluence/display/ZOOKEEPER/UsingEclipse? Thank you Ibrahim >-Original Message- >From: Flavio Junqueira [mailto:fpjunque...@yahoo.com.INVALID] >Sent: 13 April 2015 13:34 >To: user@zookeeper.ap

Adding change to ZK

manage to find the code from the Eclipse using SVN Wizard (checkout https://svn.apache.org/repos/asf/zookeeper/trunk/ zookeeper-trunk), then I got the project packages organized like (see attachment) <http://zookeeper-user.578899.n2.nabble.com/file/n7580959/package_explorer.png> Thank you I

Adding change to ZK code

manage to find the code from the Eclipse using SVN Wizard (checkout https://svn.apache.org/repos/asf/zookeeper/trunk/ zookeeper-trunk), then I got the project packages organized like (see attachment) <http://zookeeper-user.578899.n2.nabble.com/file/n7580958/package_Explorer.jpg> Thank you I

RE: Recommended JDK

Hi Guy, I've used jdk1.7.0.51 with ZK 3.4.6 for about an year and half. It works really well and I've noticed any problems. Ibrahim >-Original Message- >From: Guy Laden [mailto:guy.la...@gmail.com] >Sent: 26 March 2015 17:20 >To: zookeeper-u...@hadoop.apache.org

RE: New Zookeeper Installation - connection refused

IPs (for each 5 machines), and leave the ports the same. This how I do when I run 5 ZK servers. Ibrahim -Original Message- From: penguin [mailto:barry.barn...@wellsfargo.com] Sent: Tuesday, February 03, 2015 06:47 م To: zookeeper-u...@hadoop.apache.org Subject: New Zookeeper Installation

RE: Failover when one node fails to write on the disk?

Yes, correct. Ibrahim -Original Message- From: Sékine Coulibaly [mailto:scoulib...@gmail.com] Sent: Thursday, January 08, 2015 12:36 م To: user@zookeeper.apache.org Subject: Re: Failover when one node fails to write on the disk? Ibrahim, So, the minimum number of zk nodes is 5, not

RE: Failover when one node fails to write on the disk?

Oh yeah, I am really sorry. I thought there is 2 nodes crashes instead of one. You are absolutely right Jeff. Sorry once again Ibrahim -Original Message- From: Jeff Potter [mailto:jpotter-zookee...@codepuppy.com] Sent: Thursday, January 08, 2015 01:17 م To: user@zookeeper.apache.org

RE: Failover when one node fails to write on the disk?

running because it lacks of the quorum (majority). Ibrahim -Original Message- From: Benjamin Jaton [mailto:benjamin.ja...@gmail.com] Sent: Wednesday, January 07, 2015 10:34 م To: user@zookeeper.apache.org Subject: Failover when one node fails to write on the disk? Using zookeeper 3.4.5

RE: Question about the two-phrase commit

Thank you Alexander. I get it Ibrahim From: Alexander Shraer-2 [via zookeeper-user] [mailto:ml-node+s578899n7580745...@n2.nabble.com] Sent: Wednesday, January 07, 2015 12:29 ص To: Ibrahim El-sanosi (PGR) Subject: Re: Question about the two-phrase commit Edward is totally right - if the

RE: Question about the two-phrase commit

. Ibrahim From: Edward Carter [via zookeeper-user] [mailto:ml-node+s578899n7580741...@n2.nabble.com] Sent: Tuesday, January 06, 2015 11:53 م To: Ibrahim El-sanosi (PGR) Subject: Re: Question about the two-phrase commit If an operation is logged to disk on a minority of servers, then what happens

RE: Question about the two-phrase commit

. Which one of both statement is correct? Thank you Ibrahim From: Alexander Shraer-2 [via zookeeper-user] [mailto:ml-node+s578899n7580740...@n2.nabble.com] Sent: Tuesday, January 06, 2015 11:51 م To: Ibrahim El-sanosi (PGR) Subject: Re: Question about the two-phrase commit Yes, if the

Re: Question about the two-phrase commit

EADER message acts as the accept and UPTODATE message acts as commit). Am I right? Thank you Ibrahim -- View this message in context: http://zookeeper-user.578899.n2.nabble.com/Question-about-the-two-phrase-commit-tp7580725p7580739.html Sent from the zookeeper-user mailing list archive at Nabble.com.

Re: Question about the two-phrase commit

ransaction replicated in memory even the transaction has not got enough majority? Thank you Ibrahim -- View this message in context: http://zookeeper-user.578899.n2.nabble.com/Question-about-the-two-phrase-commit-tp7580725p7580736.html Sent from the zookeeper-user mailing list archive at Nabble.com.

Re: Exception In Zookeeper Log

Hi, Can you explain your cluster environment? Ibrahim -- View this message in context: http://zookeeper-user.578899.n2.nabble.com/Exception-In-Zookeeper-Log-tp7580722p7580723.html Sent from the zookeeper-user mailing list archive at Nabble.com.

Re: Why ZK uses TCP instead of UPD (IP multicast)?

Whereas, to achieve same thing using RUDP (IP multicast), here leader only needs to send one messages (one outgoing packets), as a result, it reduces the network traffic. Thank you Ibrahim -- View this message in context: http://zookeeper-user.578899.n2.nabble.com/Why-ZK-uses-TCP-instead

New join server while leader just sends proposal?

scenario correct (may happen)? If not how does Zookeeper deal with such scenario? Thank you Ibrahim -- View this message in context: http://zookeeper-user.578899.n2.nabble.com/New-join-server-while-leader-just-sends-proposal-tp7580714.html Sent from the zookeeper-user mailing list archive at

Why ZK uses TCP instead of UPD (IP multicast)?

Hi folks, I am just wondering why ZK uses TCP instead of UDP (IP multicast). Is there any specific reason of using TCP and not UDP (IP multicast)? Note that I meant by UDP is reliable UDP protocol, not just UDP. Thank you Ibrahim -- View this message in context: http://zookeeper-user

Re: Zookeeper myid not showing up in log4j logs

erver is connected to ZK. Ibrahim -- View this message in context: http://zookeeper-user.578899.n2.nabble.com/Zookeeper-myid-not-showing-up-in-log4j-logs-tp7580710p7580713.html Sent from the zookeeper-user mailing list archive at Nabble.com.

Re: Suggestion

Thank you Craig and Flavio, I'll take you suggestion into account Ibrahim -- View this message in context: http://zookeeper-user.578899.n2.nabble.com/Suggestion-tp7580694p7580699.html Sent from the zookeeper-user mailing list archive at Nabble.com.

Suggestion

type of operations (get, create, delete, exist, mixed (get and create), sync, and Async, different max batch size, etc). I can help with such test as I have done a huge number of testing. What do you think? Thank you Ibrahim -- View this message in context: http://zookeeper-user.578899.n2

Zookeeper don't use Nagle algorithm>

Hi folks, I just want to ask, why Zookeeper doesn't use Nagle's algorithm. I think it may improve the throughput when we activate Nagle's algorithm as it bundles the small messages? Thanks Ibrahim -- View this message in context: http://zookeeper-user.578899.n2.nabble.com/

Latency when the snapshot is taking

hen the latency time that comes between snapshot start and end time, is the latency when the snapshot is taking. Do you have efficient way to find my goal, or do you think my idea is good enough? Thank you Ibrahim -- View this message in context: http://zookeeper-user.578899.n2.nabble.com/Lat

Zookeeper doesn't allow to reset

zookeeper. Secondly, when I am trying to establish new connection (client) with above state. It gives me an error like "Error occurred during initialization of VM java.lang.OutOfMemoryError: unable to create new native thread" Can anyone help me with such issues? Thank you Ibrahim --

Re: Best strategy to test Zookeeper

. Also I am ussing Asunc mode. Can anyone share knowledge? Cheers, Ibrahim -- View this message in context: http://zookeeper-user.578899.n2.nabble.com/Best-strategy-to-test-Zookeeper-tp7580580p7580592.html Sent from the zookeeper-user mailing list archive at Nabble.com.

Re: Best strategy to test Zookeeper

Hi Alexander, I was having problem with running the ant from command line, but As soon as I set up the environment variable, I am making progress. I'll inform you what happen with me during the test. I will post any issue that I face during running the test steps, in order to be useful for anyone.

Re: Best strategy to test Zookeeper

ome line code in Zookeeper classes (FinalRequestProcessor.java and ServerStats.java). Thank a lot Alexander Ibrahim -- View this message in context: http://zookeeper-user.578899.n2.nabble.com/Best-strategy-to-test-Zookeeper-tp7580580p7580586.html Sent from the zookeeper-user mailing list archive at Nabble.com.

Re: Best strategy to test Zookeeper

d to catch end time. Can you give me hint with this? Thanks alot Ibrahim -- View this message in context: http://zookeeper-user.578899.n2.nabble.com/Best-strategy-to-test-Zookeeper-tp7580580p7580584.html Sent from the zookeeper-user mailing list archive at Nabble.com.

Best strategy to test Zookeeper

minutes are elapsed because of Async mode. Can you share knowledge? How can I solve this problem? Can anyone advise me for best strategy of testing and measuring the performance of write workload in Zookeeper? Ibrahim -- View this message in context: http://zookeeper-user.578899.n2.nabble.com

Re: Zookeeper version in github?

Hi Wendy, Its name is release-3.4.6. Thank you -- View this message in context: http://zookeeper-user.578899.n2.nabble.com/Zookeeper-version-in-github-tp7580557p7580579.html Sent from the zookeeper-user mailing list archive at Nabble.com.

Re: Zookeeper version in github?

Yes, I git it. Also we can get all the pervious versions from this link http://svn.apache.org/repos/asf/zookeeper/. Great people. Thank you a lot Ibrahim Sent from my iPhone On 10 Nov 2014, at 01:19, Wendy Smoak mailto:wsm...@gmail.com>> wrote: On Sun, Nov 9, 2014 at 8:03 PM, Ibra

RE: Zookeeper version in github?

Wendy Ibrahim -Original Message- From: Wendy Smoak [mailto:wsm...@gmail.com] Sent: Monday, November 10, 2014 12:31 ص To: user@zookeeper.apache.org Subject: Re: Zookeeper version in github? On Sun, Nov 9, 2014 at 7:08 PM, Ibrahim wrote: > So can anyone tell me what is the ve

Zookeeper version in github?

. So can anyone tell me what is the version of zookeeper that is located in the Github? Thank you Ibrahim -- View this message in context: http://zookeeper-user.578899.n2.nabble.com/Zookeeper-version-in-github-tp7580557.html Sent from the zookeeper-user mailing list archive at Nabble.com.

RE: Latency in asynchronous mode

throughput). To conclude, such high latency (392 ms) causes by batching. As example, the time 392 generates by subtracting the response and submitted time of r1 (request 1). Does this make sense? Looking forward to hearing from you. Thank you Ibrahim The interesting thing is the min values of the

RE: Latency in asynchronous mode

in async mode, the Zookeeper groups each 1000 requests and fsync them into disk once and then complete the process. Therefore, the latency affect by fsync, because the request will effect by its groups commit, that is why we see a large latency (392). I hope this makes sense to you Ibrahim

RE: ZooKeeper Meetup

Hi Hongchao, I live in UK. I am unable to come. So, is it possible to watch the event online? Thank you -Original Message- From: Hongchao Deng [mailto:hd...@cloudera.com] Sent: Thursday, October 30, 2014 05:13 م To: Jordan Zimmerman Cc: Rakesh R; Justin Kestelyn; user@zookeeper.apache.o

RE: How to run a simple Java Client

-1.6.1.jar:./slf4j-log4j12-1.6.1.jar LearnerLeaderACK/*.java java -cp .:./log4j-1.2.16.jar:./slf4j-api-1.6.1.jar:./slf4j-log4j12-1.6.1.jar LearnerLeaderACK/StartLeader Ibrahim -Original Message- From: Churu Tang [mailto:ct...@rubiconproject.com] Sent: Wednesday, October 29, 2014 08:46 م To

RE: Strange behaviour in Leader election

Yes, there is no problem with the log. Now I understand how the leader is elected and why Zookeeper elects server number 3. It is because it has the highest myid number. Thanks Ibrahim From: Flavio Junqueira-2 [via zookeeper-user] [mailto:ml-node+s578899n7580489...@n2.nabble.com] Sent

Re: Strange behaviour in Leader election

Yes, the attachment shows the logs for both leader and follower. LeaderLog.log <http://zookeeper-user.578899.n2.nabble.com/file/n7580488/LeaderLog.log> FollowerLog.log <http://zookeeper-user.578899.n2.nabble.com/file/n7580488/FollowerLog.log> Thank you Ibrahim -- View thi

RE: Strange behaviour in Leader election

Hi Flavio, Do you mean by “making the configuration and the server logs available” I try not to remove them between two runs? Sorry for misunderstanding Ibrahim From: Flavio Junqueira-2 [via zookeeper-user] [mailto:ml-node+s578899n7580485...@n2.nabble.com] Sent: Tuesday, October 28, 2014 04

Re: Strange behaviour in Leader election

Hi Rakesh , Flavio Yes, your answers make sense for me. Because I've found out that the server that has the higher myid (among the first three start servers ) become the leader. Thank you to all guys. Ibrahim -- View this message in context: http://zookeeper-user.578899.n2.nabbl

Re: Strange behaviour in Leader election

as we have a quorum (at least 3 server). Thank you Ibrahim -- View this message in context: http://zookeeper-user.578899.n2.nabble.com/Strange-behaviour-in-Leader-election-tp7580475p7580483.html Sent from the zookeeper-user mailing list archive at Nabble.com.

Re: Strange behaviour in Leader election

Hi Mingtao, In fact, I didn't configure the election algorithm at all. I just use the default one that comes with Zookeeper version 3.4.6. Cheers, Ibrahim -- View this message in context: http://zookeeper-user.578899.n2.nabble.com/Strange-behaviour-in-Leader-election-tp7580475p75

Strange behaviour in Leader election

Ibrahim -- View this message in context: http://zookeeper-user.578899.n2.nabble.com/Strange-behaviour-in-Leader-election-tp7580475.html Sent from the zookeeper-user mailing list archive at Nabble.com.

RE: Latency in asynchronous mode

Michael, Ok, In part, the question is what are you actually seeing when you look at the numbers. Which numbers do you mean? Thank you From: Michael Segel [via zookeeper-user] [mailto:ml-node+s578899n7580471...@n2.nabble.com] Sent: Sunday, October 26, 2014 12:40 ص To: Ibrahim El-sanosi (PGR

RE: Latency in asynchronous mode

one of the user. Also it is useful to follow the people replay to my question in order to become more familiar. Thank you Ibrahim -Original Message- From: Michael Segel [mailto:michael_se...@hotmail.com] Sent: Saturday, October 25, 2014 08:06 م To: user@zookeeper.apache.org Subject

RE: Latency in asynchronous mode

trying to say at min instead at max. In your scenario, if we multiply 5ms by 200 ops , we will get 1000 (second), meaning that 200 ops per second. I may be wrong or missing some things, can you correct me if so? I look forward to hearing from you. Ibrahim From: kishore g [via zookeeper-user

RE: Latency in asynchronous mode

d using async and even added additional features such a grouping of multiple requests on the client side, grouping of notifications etc. Hope this helps. thanks, Kishore G On Thu, Oct 23, 2014 at 3:31 PM, Ibrahim El-sanosi (PGR) < i.s.el-san...@newcastle.ac.uk> wrote: > WOW Kisho

RE: Latency in asynchronous mode

because my work focus in latency. Am I right? Thank you Ibrahim -Original Message- From: kishore g [mailto:g.kish...@gmail.com] Sent: Thursday, October 23, 2014 08:14 م To: user@zookeeper.apache.org Subject: Re: Latency in asynchronous mode Async api is zookeeper is a way to achieve

RE: Latency in asynchronous mode

and logical result I should take into account? It seems that the above result measures latency in Async mode request by request, whereas the result using the stat command measures batch multiple requests. Thank you a lot Ibrahim From: Henry May [via zookeeper-user] [mailto:ml-node+

RE: Latency in asynchronous mode

that it is not possible to measure the latency per operation because the transaction log will (fsync) batch multiple requests in one fsync. Can you correct me if the above is not true? Thank you a lot Ibrahim -Original Message- From: Alexander Shraer [mailto:shra...@gmail.com] Sent

RE: Latency in asynchronous mode

g the latency. Thank you for your attention I look forward to hearing from you Ibrahim -Original Message- From: Rakesh Radhakrishnan [mailto:rakeshr.apa...@gmail.com] Sent: Thursday, October 23, 2014 03:58 م To: user@zookeeper.apache.org Subject: Re: Latency in asynchronous mod

RE: Latency in asynchronous mode

, "Ibrahim" wrote: > Hi folks, > > I am testing ZooKeeper latency in Asynchronous mode. I am sending > update > (write) requests to Zookeeper cluster that consists of 5 physical > Zookeeper. > > So, when I run the stat command I get high latency like: > Latency

Latency in asynchronous mode

min/avg/max: 7/22/61 Latency min/avg/max: 7/30/65 Note that the latency measures in milliseconds. I look forward to hearing from you. Ibrahim -- View this message in context: http://zookeeper-user.578899.n2.nabble.com/Latency-in-asynchronous-mode-tp7580446.html Sent from the zookeeper

RE: How to read zookeper's logs

Hi Evelina, I don't use LogFormatter to read a log file, instead I use my own code to read the log. If you want my code I can send it to you. And if LogFormatter works with you, let me know, I want to try it. Cheers Ibrahim >-Original Message- >From: Evelina

Measure latency in cluster consists of 5 machines.

I want to measure the latency in cluster consist of 5 servers. Is the four latter word like "stat" reliable to measure the latency? Also can anyone explain how "stat" calculate the latency? Thank you Ibrahim -- View this message in context: http://zookeeper-user.578899.n

RE: Observer write behavior with partition

Yes Jordan, in the leader class, there is collection type called ConcurrentMap aim to store the write request details. If you want to look at it, it is named outstandingProposals in Leader class. Cheers, Ibrahim From: Jordan Zimmerman-3 [via zookeeper-user] [mailto:ml-node+s578899n7580372

Re: Observer write behavior with partition

, you are right. Returned observer will see the 3 writes and any missed update requests. This achieves when first the observer reconnect to the ZooKeeper cluster, the leader will send /diff / (any missed writes) to make the observer up-to-date. Regards, Ibrahim -- View this message in context

does log to disk effect the throuhput and latency

I am modified Zookeeper in order to reduce the load on the leader (trying new approach). I add extra code and only one new class to implement my approach. In the extra code, I add some log.info. When I compare modified Zookeeper (my approach) with original zookeeper, my performance is not better t