Application Stretch Physical Memory ISD Broker Stretch

Recent developments in the area of memory management have focused on reducing the eeects of the disk latency problem. These developments include the use of a compressed cache and the utilisation of memory on remote hosts. Further developments in the areas of Quality of Service (QoS) provision and user-level virtual memory have provided the impetus for a more spe-cialised run-time environment that is tailored for each individual application: an application-centric approach. Improvements on the disk latency problem have tended to be system-wide, i.e. all applications must participate in the scheme whether they would beneet or not. A compressed caching scheme is presented here as part of a new memory hierarchy that incorporates QoS for page-fault handling. The new hierarchy is executed within the application domain and utilises information about the application to best suit its individual needs. Issues in providing a compressed cache within this new memory hierarchy are presented here. It is shown that the performance of a compressed cache is aaected by many diierent factors, not counting the actual application itself, and what connguration is best for one application is not necessarily best for others. It is further demonstrated that, in conjunction with a remote paging scheme on a slow network, the compressed cache can achieve up to a factor of 3 speedup over the local disk.