Thank you, Amjad for your response. Because of our Application’s dependence on DataSources that are strictly private, it would not be useful to send you our Application.class. We are able to consistently reproduce this OutOfMemoryError event, however we reproduce it asynchronously. We are extremely impressed with what Aspose delivers to us: A Java API for rendering and aggregating MS-PPT slides. But, the run-time demand that we place on the Aspose API to render these complex financial quantitative views may be excessive. Even when we boot the Java VM with multi-GB heap capacities, we still risk encountering OutOfMemoryError . These OOME events cause the JVM to halt, breaking our SLA with banking stakeholders.
An open question for Aspose: To alleviate the necessity for Aspose end-users (like us) to necessarily install hardware RAM > 16gb (and boot their Java VMs high GB heap capacities ), would Aspose consider providing support for storing Aspose objects in off-heap JCACHE storage? This, of course, means Aspose would need to integrate a JCACHE provider (that supports off-heap capability) into their solution. Some of the JCACHE providers that support off-heap storage of objects include Hazelcast and Oracle Coherence.
Again, to accommodate the extremely complex, quantitative views we render for our clients, would Aspose consider providing support for storing Aspose objects in off-heap JCACHE storage?