Self-timed interface for S/390 I/O subsystem interconnection

نویسندگان

  • Joseph M. Hoke
  • Paul W. Bond
  • Tin-chee Lo
  • Frank S. Pidala
  • Gary Steinbrueck
چکیده

A high-speed interface has been designed for interconnection of the S/390 I/O subsystem to the IBM S/390 G5 processor. The self-timed interface (STI) provides high bandwidth, greater communication distances, and simpler timing within the S/390 servers than traditional interfaces. The STI communicates between the memory bus adapter and the expanded S/390 I/O subsystem, which now includes the new S/390 fiber channel offering (FICON) and other network-based protocols such as ATM, Fast Ethernet, and Gigabit Ethernet. Also new for G5 is the use of STI for the integrated cluster bus (ICB), providing direct links among multiple G5 servers. The STI communicates over cables up to ten meters in length at a clock frequency of 167 MHz. Data is sent at twice the clock frequency (333 MB/s). The hardware implementation comprises specially designed logic macros, differential drivers and receivers, and the cables and connectors. The receive macro accommodates up to three bittimes of skew by retiming each data bit of the interface to the transmitted clock. This paper describes the STI logic, the characteristics of the link, and the transmission and reception of the data (and clock). Introduction The self-timed interface (STI) was first introduced with the third generation of S/390* CMOS servers (G3) to satisfy the increasing I/O bandwidth requirements of these improving large systems. As the performance of the processors increased, it became apparent that in order for IBM to be successful with its new line of CMOS servers, the bandwidth and connectivity of the I/O subsystem had to scale along with the processor. STI was developed to satisfy those requirements. Figure 1 is a block diagram of an S/390 CMOS machine showing STI usage. The STI provides the interface between the memory bus adapters (MBAs) and the traditional I/O bridge chips (FIBBs), thus enabling legacy I/O such as parallel channels, ESCON*, and intersystem channels (ISC) for machines prior to G5. Today, with the delivery of the fifth generation of S/390 CMOS servers (G5), the role of the STI is further expanded as the primary interconnection method for the I/O subsystem. The STI is used for the direct connection of the new S/390 fiber channel offering (FICON*), as well as other network-based protocols such as ATM, Fast Ethernet, and Gigabit Ethernet, to the I/O subsystem through a PCI bridge chip. Also new with G5 is the ability to create direct communication links, using the STI, between other G5 servers via the Integrated Cluster Bus (ICB) [1].

برای دانلود 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

The structure of chips and links comprising the IBM eServer z990 I/O subsystem

chips and links comprising the IBM eServer z990 I/O subsystem E. W. Chencinski M. J. Becht T. E. Bubb C. G. Burwick J. Haess M. M. Helms J. M. Hoke T. Schlipf J. M. Turner H. Ulland M. H. Walz C. H. Whitehead G. Zilles The performance of large servers is to a high degree determined by their I/O subsystems. In the z990 server, nearly all of the components in the I/O path have been considerably i...

متن کامل

Hierarchical indexing data structure method for verifying the functionality of the STI-to-PCI bridge chips of the IBM eServer z900

The IBM eServer z900 has an overall system I/O bandwidth which is three times that of IBM S/390 G5/G6 servers, necessitating the use of Self-Timed-Interface (STI)-to-PeripheralComponent-Interface (PCI) bridge chips to exploit this bandwidth. The chips are used to form a layer between the networking attachments of the z900 and its main storage complex. The layer adapts the high-speed point-to-po...

متن کامل

An FPGA Prototype for a Multiplierless FIR Filter Built Using the Logarithmic Number System

Self-Timed FPGA Systems p. 21 XC6200 Fastmap Processor Interface p. 36 The Teramac Configurable Computer Engine p. 44 Telecommunication-Oriented FPGA and Dedicated CAD System p. 54 A Configurable Logic Processor for Machine Vision p. 68 Extending DSP-Boards with FPGA-Based Structures of Interconnection p. 78 High-Speed Region Detection and Labeling Using an FPGA Based Custom Computing Platform ...

متن کامل

Processor Interconnection

A multiprocessor system with a large number of nodes can be built at low cost by combining the recent advances in high capacity channels available through optical ber communication. A highly fault tolerant system is created with good performance characteristics at a reduction in system complexity. The system capitalizes of the self-routing characteristic of wavelength division multiple access t...

متن کامل

On Hybrid Synthesis for Hierarchical Structured Petri Nets

We propose a hybrid method for synthesis of hierarchical structured Petri nets. In a top-down manner, we decompose a system into a set of subsystems at each level of abstraction, each of these is speciied as a blackbox Petri net that has multiple inputs and outputs. We stipulate that each subsystem satisses the following I/O constraints: (1) At any instance of time, at most one of the inputs ca...

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

برای دانلود 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

عنوان ژورنال:
  • IBM Journal of Research and Development

دوره 43  شماره 

صفحات  -

تاریخ انتشار 1999