RE: Did anyone write custom Affinity function?
You could write a custom affinity function, and some people do, but as far as I can see you don’t need it. You just chose a poor affinity key. You need to have MANY affinity keys, much more than there are partitions, and have MANY partitions, much more than nodes. That will make sure that the default affinity function distribute data properly. But more importantly that will make sure that your system will scale well. If you have number of groups equal to the number of nodes than you can’t just increase the number of nodes to scale – you need to change your data model as well. To scale properly you need to have your data model work with different number of nodes. FYI Ignite used to have a different affinity function that always distributed partitions evenly. It had some issues and was eventually replaced and removed, although people do try to bring it back time to time. See http://apache-ignite-developers.2346864.n4.nabble.com/Resurrect-FairAffinityFunction-td19987.html Thanks, Stan From: ashishb008 Sent: 19 декабря 2018 г. 9:09 To: user@ignite.apache.org Subject: Re: Did anyone write custom Affinity function? Yeah, we were planning to increase group IDs. Did anybody write custom Affinity function? If it is already written that will be helpful to us. -- Sent from: http://apache-ignite-users.70518.x6.nabble.com/
Re: Did anyone write custom Affinity function?
Yeah, we were planning to increase group IDs. Did anybody write custom Affinity function? If it is already written that will be helpful to us. -- Sent from: http://apache-ignite-users.70518.x6.nabble.com/
Re: Did anyone write custom Affinity function?
Hi! 3 different affinity keys on 3 nodes is not enough, it's a hash code that is used so you may end up with 2 groups (or even all 3) on one node, if you need it to work that way you will need to create your own affinity function. Any way you might be able to increase the number of group id's in some way ? if you for example had 30 group id's you would have a much better distribution of the entries. Mikael Den 2018-12-18 kl. 15:14, skrev ashishb008: Hello, As of now, we have 3 nodes. We use group ID as affinity key, and we have 3 group IDs (1, 2 and 3). And we limit cache partitions to group IDs. Overall nodes=group IDs=cache partitions. So that each node have equal number of partitions. But it doesn't distribute cache partitions across the nodes. What it does, put 2 partitions on one node, 1 partition on another node, and nothing on remaining one. Thanks -- Sent from: http://apache-ignite-users.70518.x6.nabble.com/
Re: Did anyone write custom Affinity function?
Hello, As of now, we have 3 nodes. We use group ID as affinity key, and we have 3 group IDs (1, 2 and 3). And we limit cache partitions to group IDs. Overall nodes=group IDs=cache partitions. So that each node have equal number of partitions. But it doesn't distribute cache partitions across the nodes. What it does, put 2 partitions on one node, 1 partition on another node, and nothing on remaining one. Thanks -- Sent from: http://apache-ignite-users.70518.x6.nabble.com/
Re: Did anyone write custom Affinity function?
Hi! How many nodes do you have and what are you using for the affinity key ? there is of course a small chance that the distribution is not even but in most cases it will work fine as long as it has enough different values to choose from, but it depends a little bit on what you use for affinity key, say you have 10 nodes and you use a "customer id" as affinity key and you only have 8 customers it will not work of course. Mikael Den 2018-12-18 kl. 06:24, skrev ashishb008: I want all nodes in a cluster to have equal number of cache partitions. With default Affinity function it is not happening, why default does not use all nodes in a cluster? Will it be okay to write custom Affinity function? And what will we lose doing so? Did anyone write custom Affinity function? -- Sent from: http://apache-ignite-users.70518.x6.nabble.com/
