Hitachi

In-Memory Data Grid Hitachi Elastic Application Data Store


7.5.1 Specifying the locations of EADS servers

You use the eads.node.EADS-server-ID.position parameter to specify the location in the cluster where the specified EADS server is to be placed. Specifying the locations of EADS servers (hash values) enables you to distribute the workload.

If locations are not specified, the EADS servers are distributed evenly to logical locations in the cluster.

The placement of data (number of data items, data usage amounts, or data access counts) might become uneven depending on the nature of keys and applications. As a result, a large load might be placed on some of the EADS servers, resulting in poor response speeds and a shortage of resources.

Before you specify locations for the EADS servers, check the statistics (KeyCount, UsedMemorySize, and RequestCount) and analyze the intended locations.

Figure 7‒2: Establishing a location for an EADS server

[Figure]

For details about data distribution, see 2.5 Data distribution by consistent hashing.