Performance comparison of Primary Vs Secondary Indexes

Hi Am using 2.3 Doing a POC with 2 caches each having 10M records. My cluster configuration is, 3 server nodes and one client node. When I do the PT on primary key thru Rest API like this, I could TPS of around 30K and above. http://10.144.114.115:8080/ignite?cmd=get&key=P10007&cacheName

Re: Ignite in docker (Native Persistence)

Hello! Yes, sure! I'll investigate this question and update our doc. On Thu, Dec 7, 2017 at 9:57 AM, Denis Magda wrote: > Nick, > > As one of Ignite docker maintainers, could you please investigate one how > to map Ignite persistence to docker volumes: > https://docs.docker.com/engine/admin/vol

Re: How IGFS URIs work

OK, I think you are missing this part : your Spark compute nodes that you say have no local Ignite still must have Ignite configuration and Ignite dependencies on the Hadoop's class path. To make your Spark no

Re: ClassNotFoundException when using IgniteHadoopIgfsSecondaryFileSystem

I just replied you on another thread with details on what you need to do to connect to a remote IGFS cluster. You still need Ignite configuration and Ignite-Hadoop dependencies on your spark nodes to connect to r

e-mail address change not effective

Hello, I changed e-mail address and have performed the following actions: - I confirmed the address change - I sent unsubscribe e-mail from my old address - I sent subscribe e-mail from my new address I keep getting e-mails at my old address... -- Sent from: http://apache-ignite-users.70518.x

CacheStoreAdapter loadCache overwrite existing values

Im using CacheStoreAdapter to load my tables from a database into ignite cache. I need to update the cache periodically during runtime (no downtime allowed). Can I just call `.loadCache(null)` again on my existing cache? Will existing entries be overwritten (I need them to get overwritten). --

Re: Re:Re: Re:Re: index about jdbc thin mode

Hi, Your configuration perfectly works on my side. Make sure that you are connecting to the server node with an appropriate configuration of indexes. I'm afraid, that without a complete reproducer it's not possible to figure out the reason. Kind regards, Alex -- Sent from: http://apache-ignit

Use custom Data Region or custom Cache for IgniteAtomicReference - Ignite 2.3

Hi all, I would like to ask, if there is a way to use custom DataRegionConfiguration (other than default) for storing IgniteAtomicReferences. I would like to use Ignite's 2.3 feature where I can turn on/off Persistent per Cache (via DataRegionConfiguration). I would like to enable persistent only

Re: Use custom Data Region or custom Cache for IgniteAtomicReference - Ignite 2.3

Hi Krzysztof! You can do this via *AtomicConfiguration#setGroupName*. For example: AtomicConfiguration cfg = new AtomicConfiguration().setGroupName("atomicRefCacheGroup"); ignite.atomicReference("atomicRef", cfg, "initValue", true); where "atomicRefCacheGroup" cache group that has your DataReg

Re: Problem with loading data

Dear; After I did run the server code locally in debug mode, I found the problematic query: SELECT id, job_id, applicant_id, bucket, bucket_id, STATUS, ORDER, is_new, candidate_archived, candidate_archived_date,

Re: Ignite in docker (Native Persistence)

Nick, Excellent, here is a ticket! Please reassign it on me once you complete with the doc: https://issues.apache.org/jira/browse/IGNITE-7132 — Denis > On Dec 7, 2017, at 3:57 AM, Nikolai Tikhonov wrote: > > Hello! > > Yes, sure! I'll investigate this question and update our doc. > > On Thu

Re: Problem with loading data

There is a cache configuration property "sqlEscapeAll" that might address your issue. Set it to "true" to make Ignite put all SQL query fields inside double quotes.

Re: using JDBC with Ignite cluster, configured with persistent storage

Here is what happened: 1) I've configured standalone cluster of 5 nodes with persistent storage enabled in data storage part. Following is the section of my configuration file:

Re: How IGFS URIs work

Just reviewed Remote IGFS TCP client implementation - you will get load balancing if you use "igfs://myIgfs@host:port/" format but you will not get high availability. That means your connection would fail if the "host" goes down. So you have two options: 1. Hight availability IGFS client <

Affinity - Join query on the collocated data taking 90 seconds

Hi AM using 2.3 Have 2 caches Customer - PartyId is the Primary Key Account - AccountId is the primary key and also has another column called PartyId While storing the Account data, I am using AffinityKey, so that my below join query works since the data is collocated, I could get the result for

Re: Performance comparison of Primary Vs Secondary Indexes

Hello! It looks as expected behaviour. In the first request you use IgniteCache API. In this case Apache Ignite knows which node in a cluster have a data and fetch a entry from it. Only one request and one responce. But in the second request Ignite does more action. In the first, Apache Ignite pa

Re: Performance comparison of Primary Vs Secondary Indexes

Hi Nikolay Can it get deteriorated further if we have more nodes in the cluster ? If so, performance of secondary index is inversely proportional to the number of nodes Thanks Naveen -- Sent from: http://apache-ignite-users.70518.x6.nabble.com/

Re: e-mail address change not effective

Hello, Did you receive approved message when sent from old address? Any way look at steps listed below, might be it will helpful for you. To unsubscribe from the user mailing list send a letter to user-unsubscr...@ignite.apache.org with a word "Unsubscribe" without quotes as a topic. If you have

How can we use the Discovery URL with SQLLINE which is both load balanced and Fault tolerent

Hi I am using 2.3, have 3 nodes in my cluster, This is how the config XML entries look like, all 3 nodes have below entry

Re: Performance comparison of Primary Vs Secondary Indexes

Seems that there is no direct dependence of influence of the number of nodes on performance. It strong depends from your case. Any way, you need to do experiment for your case and analyze results. On Thu, Dec 7, 2017 at 6:07 PM, Naveen wrote: > Hi Nikolay > > Can it get deteriorated further if w

Re: Redis problem with values > 8kb

Hi Wolfram, Ups, it seems I lost your answer in email flood. I will check PHP sample at my side tomorrow and return back. Thank you, Alexey -- Sent from: http://apache-ignite-users.70518.x6.nabble.com/

Re: CacheStoreAdapter loadCache overwrite existing values

Hello, > Can I just call `.loadCache(null)` again on my existing cache? CacheStore.loadCache() is generally used for loading the cache on startup, but it can be used at any point after the cache has been created. > Will existing entries be overwritten (I need them to get overwritten). No, existi

Re: How can we use the Discovery URL with SQLLINE which is both load balanced and Fault tolerent

Hi! In current implementation thin driver doesn't support these features. You can use thick driver [1] that supports them. 1. https://apacheignite.readme.io/docs/jdbc-driver#jdbc-client-node-driver On Thu, Dec 7, 2017 at 6:28 PM, Naveen wrote: > Hi > > I am using 2.3, have 3 nodes in my cluste

cluster hanged when client node put data into the caches

Hi guys. our project was stunned into this critical issue. Ignite version 2.2, 10 nodes on 5 servers. Client node consumed data from Kafka, and put them into caches. Sometimes client node was disconnected suddenly, then the cluster with 10 server nodes hanged, CPU usage on 5 servers reached almost

Re: Affinity - Join query on the collocated data taking 90 seconds

Hi, Did you create indexes for PARTY_ID fields? Any way, can you share explain for the query and also try to rewrite the query via inner join? On Thu, Dec 7, 2017 at 5:59 PM, Naveen wrote: > Hi > > AM using 2.3 > Have 2 caches > Customer - PartyId is the Primary Key > Account - AccountId is the

Re: cluster hanged when client node put data into the caches

Hello, Can you share thread dumps from all nodes? It need for further investigation. On Thu, Dec 7, 2017 at 7:41 PM, Aurora <2565003...@qq.com> wrote: > Hi guys. > our project was stunned into this critical issue. > Ignite version 2.2, 10 nodes on 5 servers. > Client node consumed data from Kafk

Re: Affinity - Join query on the collocated data taking 90 seconds

Look at there [1] how to use explain statement. 1. https://apacheignite-sql.readme.io/docs/performance-and-debugging#using-explain-statement On Thu, Dec 7, 2017 at 7:43 PM, Nikolai Tikhonov wrote: > Hi, > > Did you create indexes for PARTY_ID fields? > Any way, can you share explain for the que

Re: Affinity - Join query on the collocated data taking 90 seconds

I have created an index on partyId of Account cache, after that query is responding in milli seconds. However, my basic affinity and collation does not seems to be working. I have created records with Ids from P10100 to P10101. If you see below, some of the records are not returned, doe

Re: ClassNotFoundException when using IgniteHadoopIgfsSecondaryFileSystem

Hi Alexey, Thanks for your answers. I was finally able to fix this, I was missing some additional jars in the lib directory of the ignite installation in the worker nodes. The exception thrown was by org.apache.hadoop.conf.Configuration.getClassByName was a bit misleading because it was referring

Re: How IGFS URIs work

Thanks again Alexey, you are right, I hadn't noticed about the IGFS HA client. I'll give it a try, the embedded client shouldn't be too much overhead if I use instance types that are big enough. Thanks again for all your help! Greetings, Juan On Thu, Dec 7, 2017 at 6:41 AM, Alexey Kukushkin w

Re: Register Contentious query quit slow

Denis, Thanks for your reply and the code as well I'll try what you suggested. Thanks again BR. Jin -- Sent from: http://apache-ignite-users.70518.x6.nabble.com/

Re: cluster hanged when client node put data into the caches

@Nikolai thank you for your response. some error in client node: 2017-12-06 13:35:03.721 -ERROR 9688 [order] com.a.u.IgniteAffinityHelper : error: org.apache.ignite.IgniteClientDisconnectedException: Client node disconnected: null at org.apache.ignite.internal.GridKernalGatewayIm

Re: cluster hanged when client node put data into the caches

thread dump: Attaching to process ID 12837, please wait... Debugger attached successfully. Server compiler detected. JVM version is 25.65-b01 Deadlock Detection: No deadlocks found. -

Re: "HandshakeException: Remote node ID is not as expected" occurs when tested with 1000 clients

It seems a problem. I have checked a lot of failed and not failed spark tasks, and all failed tasks had "Cache is not configured: ignite-sys-cache" exception message, and all not failed tasks did not. And, this exception mostly occurs at the early period of job, and after running at a certain amou