Status

Current state: Done

Released: 4.5.0

Problem

It is likely that in some Bookkeeper clusters, we’ll have machines with different hardware capabilities. Due to the need for expanding capacity in a cluster, it is possible that we end up having bookies of different SKUs(No. of CPUs, memory/storage capacity, etc) in a single cluster. In such environments, we can have a policy that makes sure that bookies with smaller storage capacity don’t run out of disk space and turn read-only quickly. Instead we want the storage usage on those bookies to grow at a pace that is proportional to their capacity. Another policy could be to select bookies that are experiencing lower system load more often during ledger creation, than the ones that are experiencing higher load. Yet another approach could take each bookies network bandwidth into account and use it as a weight. 

...

1. DefaultEnsemblePlacementPolicy: Changes will be made to newEnsemble() and replaceBookie() interfaces to select the bookies based on their free disk space based weight when the configuration suggests that weight based placement is to be honored. 
2. RackawareEnsemblePlacementPolicy: Enhancements will be made to selectRandomFromRack() and selectRandomInternal() to take into account the weights of the bookies. No changes will be made to the rack selection logic. But once a rack has been selected the bookie selection within that rack will be based on disk weight. 
3. RegionAwareEnsemblePlacementPolicy: No changes are needed to the region selection logic. In this policy  once a region is selected, the bookie selection is done using RackawareEnsemblePlacementPolicy. 

1.4.2 BookieInfo

The following protocol message is being added to retrieve the BookieInfo details. This message can be extended to retrieve other information such as network bandwidth in the future.

message GetBookieInfoRequest {
    enum Flags {
        TOTAL_DISK_CAPACITY = 0x01;
        FREE_DISK_SPACE = 0x02;
     }
     // bitwise OR of Flags
     optional int64 requested = 1;
}

message GetBookieInfoResponse {
    required StatusCode status = 1;
    optional int64 totalDiskCapacity = 2;
    optional int64 freeDiskSpace = 3;
}

1.4.3 BookieInfoReader

An instance of BookieInfoReader is created to read periodically the bookie information such as free disk space, bookie load etc. This singleton is instantiated once per Bookkeeper object. It uses a SingleThreadedExecutor to query the bookie info from all the bookies. It then communicates the updated information to the ensemble policies via the updateBookieInfo() interface described above. Changes in the number of bookies in the cluster also triggers this operation. The event is triggered from BookieWatcher singleton.

...