Richard Fromm

A case for intelligent RAM

Two trends call into question the current practice of microprocessors and DRAMs being fabricated as different chips on different fab lines: 1) the gap between processor and DRAM speed is growing at...Two trends call into question the current practice of fabricating microprocessors and DRAMs as different chips on different fabrication lines. The gap between processor and DRAM speed is growing at 50% per year; and the size and organization of memory on a single DRAM chip is becoming awkward to use, yet size is growing at 60% per year. Intelligent RAM, or IRAM, merges processing and memory into a single chip to lower memory latency, increase memory bandwidth, and improve energy efficiency. It also allows more flexible selection of memory size and organization, and promises savings in board area. This article reviews the state of microprocessors and DRAMs today, explores some of the opportunities and challenges for IRAMs, and finally estimates performance and energy efficiency of three IRAM designs.

Scalable processors in the billion-transistor era: IRAM

Members of the University of California, Berkeley, argue that the memory system will be the greatest inhibitor of performance gains in future architectures. Thus, they propose the intelligent RAM or IRAM. This approach greatly increases the on-chip memory capacity by using DRAM technology instead of much less dense SRAM memory cells. The resultant on-chip memory capacity coupled with the high bandwidths available on chip should allow cost-effective vector processors to reach performance levels much higher than those of traditional architectures. Although vector processors require explicit compilation, the authors claim that vector compilation technology is mature (having been used for decades in supercomputers), and furthermore, that future workloads will contain more heavily vectorizable components.

The energy efficiency of IRAM architectures

Portable systems demand energy efficiency in order to maximize battery life. IRAM architectures, which combine DRAM and a processor on the same chip in a DRAM process, are more energy efficient than conventional systems. The high density of DRAM permits a much larger amount of memory on-chip than a traditional SRAM cache design in a logic process. This allows most or all IRAM memory accesses to be satisfied on-chip. Thus there is much less need to drive high-capacitance off-chip buses, which contribute significantly to the energy consumption of a system. To quantify this advantage we apply models of energy consumption in DRAM and SRAM memories to results from cache simulations of applications reflective of personal productivity tasks on low power systems. We find that IRAM memory hierarchies consume as little as 22% of the energy consumed by a conventional memory hierarchy for memory-intensive applications, while delivering comparable performance. Furthermore, the energy consumed by a system consisting of an IRAM memory hierarchy combined with an energy efficient CPU core is as little as 40% of that of the same CPU core with a traditional memory hierarchy.

A 300-MHz 64-b quad-issue CMOS RISC microprocessor

This 300 MHz quad-issue custom VLSI implementation of the Alpha architecture delivers 1200 MIPS (peak), 600 MFLOPS (peak), 341 SPECint92, and 512 SPECfp92. The 16.5 mm/spl times/18.1 mm die contains 9.3 M transistors and dissipates 50 W at 300 MHz. It is fabricated in a 3.3 V, four-layer metal, 0.5 /spl mu/m, CMOS process. The upper metal layers (metal-3 and metal-4), primarily used for power, ground, and clock distribution. The chip supports 3.3 V/5.0 V interfaces and is packaged in a 499-pin ceramic IPGA. It contains an 8-kbyte instruction cache; an 8-kbyte, dual-ported, data cache; and a 96-kbyte, unified, second-level, 3-way set associative, fully pipelined, writeback cache. This paper describes the circuit and implementation techniques that were used to attain the 300 MHz operating frequency.

A 433 MHz 64 b quad issue RISC microprocessor

This 9.6 M transistor quad-issue RISC microprocessor achieves greater than 500 SPECint92 (estimated) at 433 MHz. The die measures 14.5/spl times/14.4 mm/sup 2/ fabricated in a 0.35 /spl mu/m CMOS process. The chip uses split-power supplies; the core operates at 2.0 V and the external interface at 3.3 V. The chip dissipates less than 25 W.

Intelligent RAM (IRAM): the industrial setting, applications, and architectures

The goal of intelligent RAM (IRAM) is to design a cost-effective computer by designing a processor in a memory fabrication process, instead of in a conventional logic fabrication process, and include memory on-chip. To design a processor in a DRAM process one must learn about the business and culture of the DRAMs, which is quite different from microprocessors. The authors describe some of those differences and their current vision of IRAM applications, architectures, and implementations.

Software video production switcher

We have developed an application for use in the production of digital video broadcasts. This application, the Software Video Production Switcher, is an extensible application that is used to manage a set of continuous video inputs in realtime. The switcher takes these video inputs and performs user-directed actions to produce a single output video stream. These actions include simple cuts between video streams, more complex video transitions, and special video effects. The output stream can then be broadcast onto a network and received by remote viewer applications for playback.