The Cache plug-in for vRealize Orchestrator is a a plug-in which provides distributed in-memory data structures such as maps, queues sets etc.
Basic features of the plug-in:
- Distributed in-memory data structures with TTL/lease options
- Distributed unique ID generation
- Distributed locking
- Automatic auto-discovery with other vRO nodes which belong to the cluster
The Cache plug-in supports vRO 6.0 and later.
#####Using lists:
//Add a value in the default list
CacheManager.listService.add("my-value");
//Get the first value from the list
var value = CacheManager.listService.get(0);
//Clear the default list
CacheManager.listService.clear();
#####Using sets:
//Add a value in the default set
CacheManager.setService.add("my-value");
//Get the elements from the default set
var elements = CacheManager.setService.elements();
//Clear the default set
CacheManager.setService.clear();
#####Using maps:
//Put a value in the default map without TTL
CacheManager.mapService.put("my-key", "my-value");
//Put a value in the default map with a TTL of 2 minutes
CacheManager.mapService.put("my-key", "my-other-value", 2, CacheTimeUnit.MINUTES);
//Get the value from the map
var value = CacheManager.mapService.get("my-key");
#####Using queues:
//Offer a value in the default queue, waiting up to 2 minutes if necessary for space to become available
CacheManager.queueService.offer("my-value", 2, CacheTimeUnit.MINUTES);
//Offer a value in the default queue, waiting if necessary for space to become available
CacheManager.queueService.offer("my-value");
//Poll a value from the default queue
var value = CacheManager.queueService.poll();
//Put a value from the default queue, waiting if necessary for space to become available
CacheManager.queueService.put("my-value");
//Retrieve and remove the head of the queue. Throws an exception if queue is empty
val value = CacheManager.queueService.remove();
#####Using ring buffers:
//Adds an item to the tail of the ringbuffer
CacheManager.ringbufferService.add("my-value");
//Get the sequence of the head
var sequence = CacheManager.ringbufferService.headSequence();
while(true){
//Reads the item from the ringbuffer
var item = CacheManager.ringbufferService.readOne(sequence);
sequence++;
}
#####Using ID generators:
//Create a unique ID with the default generator
var id = IdGeneratorManager.idGeneratorService.newId();
//Create a unique ID with a named ID generator
var id = IdGeneratorManager.idGeneratorService.newIdForGenerator("my-id-generator");
#####Using locks:
//Acquires a lock by current workflow token id for 2 minutes
LockManager.lockService.lock(workflow.id, 2, CacheTimeUnit.MINUTES);
//Releases the lock. Will fail if invoked from another workflow
LockManager.lockService.unlock(workflow.id);
//Tries to acquire a lock if it is free within 10 seconds. Returns true if succeeded
var result = LockManager.lockService.tryLock(workflow.id, 10, CacheTimeUnit.SECONDS);
//Tries to acquire a lock for 2 minutes, if it is free within 10 seconds. Returns true if succeeded
var resut = LockManager.lockService.tryLockWithLease(workflow.id, 10, CacheTimeUnit.SECONDS, 2, CacheTimeUnit.MINUTES);
The Cache plug-in for vRO uses the following third-party libraries:
- Hazelcast - an in-memory data grid middleware, distributed under the Apache License, Version 2.0.
- Google Guava - core libraries for Java-based projects, distributed under the Apache License, Version 2.0