The AlphaServer 4100 Low-cost Clock Distribution System

نویسنده

  • Roger A. Dame
چکیده

tion system to synchronize electronic communication. The primary metric used to quantify the performance of a clock distribution system is clock skew. Synchronous systems require multiple copies (outputs) of the same clock, and clock skew is the unwanted delay between any two of the copies. In general, the lower the skew, the better the clock system. Clock skew is one of several parameters that affect bus speed. Bus length, bus loading, driver and receiver technology, and bus signal voltage swing also affect bus speed. If problems arise that jeopardize meeting timing goals, though, these additional parameters are difficult to change because of physical and architectural constraints. The DIGITAL AlphaServer 4100 clock distribution system is a compact, low-cost solution for a highperformance midrange server. The clock system provides more copies of the clock than machines in the same class typically need. The distribution system allows expansion on those module designs where more copies of the clock are needed with minimal skew. The system is based on a low-cost, off-the-shelf phase-locked loop (PLL) as the basic building block. The simple application of the PLL alone would not provide low clock skew, though. Signal integrity techniques and trade-offs were needed to manage skew throughout the system. The technical challenges were to design a low-cost system that would (1) require only a small area on the printed wiring boards (PWBs), (2) be adaptable to various speed grades (options) of CPUs, and (3) have good performance, i.e., low skew. This paper discusses the techniques used to optimize the performance of an off-the-shelf PLL-based clock distribution system.

برای دانلود رایگان متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

High Performance I/O Design in the AlphaServer 4100 Symmetric Multiprocessing System

ing system based on the Alpha 21164 64-bit RISC microprocessor. This midrange system supports one to four CPUs, one to four 64-bit-wide peer bridges to the peripheral component interconnect (PCI), and one to four logical memory slots. The goals for the AlphaServer 4100 system were fast CPU performance, low memory latency, and high memory and I/O bandwidth. One measure of success in achieving th...

متن کامل

Design and Implementation of the AlphaServer 4100 CPU and Memory Architecture

Vol. 8 No. 4 1996 The DIGITAL AlphaServer 4100 system is a symmetric multiprocessing (SMP) midrange server that supports up to four Alpha 21164 microprocessors. A single Alpha 21164 CPU chip may simultaneously issue multiple external accesses to main memory. The AlphaServer 4100 memory interconnect was designed to maximize this multiple-issue feature of the Alpha 21164 CPU chip and to take adva...

متن کامل

The AlphaServer 4100 Cached Processor Module Architecture and Design

sents the third generation of Alpha microprocessorbased, mid-range computer systems. Among the technical goals achieved in the system design were the use of four CPU modules, 8 gigabytes (GB) of memory, and partial block writes to improve I/O performance. Unlike the previous generation of mid-range servers, the AlphaServer 4100 series can accommodate four processor modules, while retaining the ...

متن کامل

Building a High-performance Message-passing System for MEMORY CHANNEL Clusters

DEVICE INTERFACE ABSTRACT DEVICE INTERFACE Figure 14 Digital MPI Architecture Table 5 MPI Latency Comparison MPI Implementation Transport Platform Latency MPICH 1.0.10 Sockets FDDI DEC 3000/800 350 ms MPICH 1.0.10 Shared Memory AlphaServer 2100 4/233 30 ms Digital MPI V1.0 MEMORY CHANNEL 1.0 AlphaServer 2100 4/233 16 ms Digital MPI V1.0 MEMORY CHANNEL 1.5 AlphaServer 4100 5/300 6.9 ms Digital M...

متن کامل

AlphaServer 4100 Performance Characterization

The performance analysis undertaken examined a number of workloads with different characteristics, including the SPEC95 benchmark suites (floating-point and integer), the LINPACK benchmark, AIM Suite VII (UNIX multiuser benchmark), the TPC-C transaction processing benchmark, image rendering, and memory latency and bandwidth tests. Note that both commercial (AIM and TPC-C) and technical/scientif...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

عنوان ژورنال:
  • Digital Technical Journal

دوره 8  شماره 

صفحات  -

تاریخ انتشار 1996