To execute shared-memory-based parallel programs efficiently, we introduce two compiler-assisted software cache schemes which are well-suited to automatic optimizations of remote communications. One scheme is a full user-level software cache (User-level Distributed Shared Memory: UDSM) and another is a page-based cache (Asymmetric Distributed Shared Memory: ADSM) which exploits TLB/MMU only in ...

We study the cache performance in a remote caching architecture. The high performance networks in many distributed systems enable a site to access the main memory of other sites in less time than required by local disk access. Remote memory is thus introduced as an additional layer in the memory hierarchy between local memory and disks. Eficient use of remote memory implies that the system cach...

We develop and examine job migration policies by considering effective usage of global memory in addition to CPU load sharing in distributed systems. When a node is identified for lacking sufficient memory space to serve jobs, one or more jobs of the node will be migrated to remote nodes with low memory allocations. If the memory space is sufficiently large, the jobs will be scheduled by a CPU-...

In the existing Fault-Tolerant Software Distributed Shared Memory (FT-SDSM) with the message logging, the logs are used only to recover the failed nodes. In our previous work, we have implemented a lightweight logging protocol, called remote logging, on the SDSM for fault tolerance, which incurs low logging overhead with a fast network and a remote memory for back-up data. In this paper, we pro...

One of the most important performance issues on NUMA systems is data locality since remote memory accesses have latencies several magnitudes higher than local memory accesses. This paper presents a tool environment targeting at tuning NUMA-based shared memory applications towards better memory locality. This tool environment comprises tools, supporting system facilities, and their interface. To...

In virtual machine systems, with the increase in the number of VMs and the demands of the applications, the main memory is becoming a bottleneck for the application performance. To improve paging performance and alleviating thrashing behavior for memory-intensive or I/O-intensive virtual machine workloads, we proposed hypervisor based remote paging, which allows a virtual machine to transparent...

One common cause of poor performance in large-scale shared-memory multiprocessors is limited memory or interconnection network bandwidth. Even well-designed machines can exhibit band-width limitations when a program issues an excessive number of remote memory accesses or when remote accesses are distributed non-uniformly. While techniques for improving locality of reference are often successful...