The memory wall (the gap between processing and storage speeds) remains a concern to computer systems designers. Caches have played a key role in hiding the performance gap by keeping recently accessed information in fast memories closer to the processor. Multi and many core systems are placing severe demands on caches, exacerbating the performance disparity between memory and processors. New m...

The RAMpage memory hierarchy is an alternative memory organization which addresses the growing CPU-DRAM speed gap, by replacing the lowest-level cache by an SRAM main memory. This paper presents some modifications to the RAMpage hierarchy. More aggressive first level cache implementations are shown to improve performance of the RAMpage model, when context switches were taken on misses to DRAM.

The number of concurrent video or audio streams a disk-based Continuous Media Server can support, is limited by the time needed for the associated data blocks to be retrieved from secondary storage. Along with the cost of RAM, these two strongly interrelated parameters determine the cost/performance metric of the Media Server. Research efforts have focused either on maximizing disk performance ...

Large scale storage servers that provide location transparent, interactive access to hundreds or thousands of concurrent, independent clients will be important components of hte furture information super-highway infrastructure. Two key requirements of such servers are as follows: support high parallelism and concurrency in data access to allow large number of access to the same or different dat...

Oblivious RAM (ORAM) garbles read/write operations by a client (to access a remote storage server or a random-access memory) so that an adversary observing the garbled access sequence cannot infer any information about the original operations, other than their overall number. This paper considers the natural setting of Oblivious Parallel RAM (OPRAM) recently introduced by Boyle, Chung, and Pass...

Given the present cost of memories and the very large storage and bandwidth requirements of large-scale multimedia databases, hierarchical storage servers (which consist of RAM, disk storage, and robot-based tertiary libraries) are becoming increasingly popular. However, related research is scarce and employs tertiary storage for storage augmentation purposes only. This work, exploiting the eve...

From the last few decades, the scaling down of CMOS devices have been taking place to achieve better performance in terms of speed, power dissipation, size and reliability. The major area of concern in today‟s CMOS technology is Data retention and leakage current reduction. SRAM (Static Random Access Memory) is memory used to store data. Conventional Static Random Access Memory (SRAM) cells suf...

Main memory is one of the major energy consumers in computing systems. In this paper, we propose a new memory organisation, called Power-Aware Cached-DRAM (PA-CDRAM) that integrates a moderately sized cache directly into a memory device. We use this cache to turn memory banks off immediately after memory accesses to reduce energy consumption. We modify CDRAM (originally optimised for performanc...

Compression has been used typically to save disk space and bandwidth. However, from a decade ago, a new family of data structures allows applying compression to save space in main memory as well. The so-called self-indexes are an example of these new structures. They can store a text in space proportional to that of the compressed version of the same text, and at the same time, they allow searc...