Research Projects

Motivation

Computing is changing dramatically, particularly for cloud-based service providers, such as Facebook, Google, Amazon, etc. These on-line service applications, such as social networking and search place unique demands on processor memory systems. In particular, these big memory applications have working data sizes several orders of magnitude beyond those found in the workloads typically used in computer systems research. Further, these applications can place varying demands on the memory system in terms of performance and persistence characteristics. Simultaneously, new, Non-Volatile Memory technologies such as PCIe attached Flash RAM, Phase Change Memory (PCM), STT-RAM and memristors, are emerging for use as augmentation to traditional DRAM main memory. These new memory technologies promise higher capacities, fast access times along with non-volatility (i.e. data retention when the power is off).  They have very different reliability characteristics requiring new coding and signal processing techniques, systems design and data management policies.  These new technologies have the potential to bridge the gaps in current processor memory systems for both data capacity and speed requirements, leading to new usage models, such as storage class memories or combined main memory and storage implementations.   These trends together argue for new memory systems architectures, designed for the challenges of big memory applications, leveraging new memory technologies together with traditional DRAM and magnetic storage, which expose their intrinsic technology characteristics to the programmer.

Memory Storage Hierarchy

System Software for NVM

Coding and Signal Processing