Distributed memory with hard disk overflow.

[Alvaro Martin]

Hello all,

I have recently begun with Hazelcast to try to solve my problem. The scenario I have is a set of threads sharing the memory space, and processing concurrently a large ammount of data. I have seen that Hazelcast does not easily support distributed persistence, but is there any way to configure a database persistence for those cases where the shared memory is not enought?. In my case, I want to use a backup in every machine, available for reading because I need all the memory to be shared with all the threads, that is, I will be using hazelcast because its syncronization abilities.

Thanks.

Álvaro.

[Andrejs Jermakovics]

Hi Alvaro,

Yes, you can configure database persistence: http://docs.hazelcast.org/docs/3.4/manual/html-single/hazelcast-documentation.html#map-persistence

We also have more code samples here https://github.com/hazelcast/hazelcast-code-samples/tree/master/distributed-map/mapstore

Reading backup data is also possible http://docs.hazelcast.org/docs/3.4/manual/html-single/hazelcast-documentation.html#read-backup-data

Hope this helps

Best,

Andrejs 

--
You received this message because you are subscribed to the Google Groups "Hazelcast" group.
To unsubscribe from this group and stop receiving emails from it, send an email to hazelcast+...@googlegroups.com.
To post to this group, send email to haze...@googlegroups.com.
Visit this group at http://groups.google.com/group/hazelcast.
To view this discussion on the web visit https://groups.google.com/d/msgid/hazelcast/1e9de4b9-9ae5-4ab4-87dd-22880ea8ac3c%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

--

--

Best,

Andrejs

[Alvaro Martin]

Hello Andrejs,

I know that I can configure persistence, and also backup read. The problem that I have is that the documentations says that there can be only one centralized database, and I want them to be distributed.

The arquitecture I have is something like:

Node A - DataBase A

Node B - DataBase B

Node C - DataBase C

So all nodes uses their database to store everything (because I want a backup-all). That means, if I make a put(someKey, someValue) I want this to be stored in DataBase A, DataBase B and DataBase C and also to be available for a fast-read in the memory of Node A, Node B and Node C. As far as I know, when I make a put(someKey, someValue) it will be stored as original in some node (let's say Node A) and as backup in both others (let's say Node B and Node C). If I configure a database persistence, it will be also persisted in that database but it will be placed only in one site (Node A, Node B, Node C, or any other Node X), but not replicated in every node.

Thanks.

Álvaro.

--
You received this message because you are subscribed to a topic in the Google Groups "Hazelcast" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/hazelcast/LW8-_0vpcTU/unsubscribe.
To unsubscribe from this group and all its topics, send an email to hazelcast+...@googlegroups.com.

To post to this group, send email to haze...@googlegroups.com.
Visit this group at http://groups.google.com/group/hazelcast.

To view this discussion on the web visit https://groups.google.com/d/msgid/hazelcast/CAF1jyT%2B6wOU5-4iVwNE4j%2BAjpB6-TPg7H6esRNPn-SV%2BNbtVjQ%40mail.gmail.com.

For more options, visit https://groups.google.com/d/optout.

--

Álvaro Martín Lozano

Móvil: 686764257

Correo: alvaro...@gmail.com

[dsukho...@gmail.com]

Hi Alvaro,

In your MapStore class you can persist cached entries in as many DBs as you like, it depends on your impleentation only.

Denis.

[Andrejs Jermakovics]

Oh right, I must have misunderstood you. The documentation is correct and currently it has to be one centralised database.

Distributed persistence is on the road map however.

To view this discussion on the web visit https://groups.google.com/d/msgid/hazelcast/CANMmWQGBJkHbVv0WF5TY9jwOFg70KCaqk-PNeA%3DPM%3DKqvjv0tQ%40mail.gmail.com.

For more options, visit https://groups.google.com/d/optout.

--

--

Best,

Andrejs

[Lukas Blunschi]

Hi Andrejs and Alvaro,

true, only one centralised database is supported, however if that "centralised" database is a distributed key-value store like Cassandra, it should work for you.

Best,

Lukas