rfc1152.txt

RFC 的详细文档！

   providing a multicast capability is critical.

   Finally, Greg Watson (HP) presented an overview of ongoing work at
   the Hewlett-Packard Bristol lab.  Their belief is that, while
   applications for high-speed networks employing supercomputers are the
   the technology drivers, the economic drivers will be applications
   requiring moderate bandwidth (say 10 Mbps) that are used by everyone
   on the network.

   They are investigating how multimedia workstations can assist
   distributed research teams - small teams of people who are
   geographically dispersed and who need to work closely on some area of
   research.  Each workstation provides multiple video channels,
   together with some distributed applications running on personal
   computers.  The bandwidth requirements per workstation are about 40
   Mbps, assuming a certain degree of compression of the video channels.
   Currently the video is distributed as an analog signal over CATV
   equipment.  Ideally it would all be carried over a single, unified
   wide-area network operating in the one-to-several Gbps range.



Partridge                                                       [Page 5]

RFC 1152                  IRSG Workshop Report                April 1990


   They have constructed a gigabit network prototype and are currently
   experimenting with uncompressed video carried over the same network
   as normal data traffic.

Session 3: Lightwave Technology and its Implications (Ira Richer, Chair)

   Bob Kennedy (MIT) opened the session with a talk on network design in
   an era of excess bandwidth.  Kennedy's research is focused on multi-
   purpose networks in which bandwidth is not a scarce commodity,
   networks with bandwidths of tens of terahertz.  Kennedy points out
   that a key challenge in such networks is that electronics cannot keep
   up with fiber speeds.  He proposes that we consider all-optical
   networks (in which all signals are optical) with optoelectronic nodes
   or gateways capable of recognizing and capturing only traffic
   destined for them, using time, frequency, or code divisions of the
   huge bandwidth.  The routing algorithms in such networks would be
   extremely simple to avoid having to convert fiber-optics into slower
   electronic pathways to do switching.

   Rich Gitlin (AT&T Bell Labs) gave a talk on issues and opportunities
   in broadband telecommunications networks, with emphasis on the role
   of fiber optic and photonic technology.  A three-level architecture
   for a broadband telecommunications network was presented.  The
   network is B-ISDN/ATM 150 (Mbps) based and consists of: customer
   premises equipment (PBXs, LANs, multimedia terminals) that access the
   network via a router/gateway, a Network Node (which is a high
   performance ATM packet switch) that serves both as a LAN-to-LAN
   interconnect and as a packet concentrator for traffic destined for
   CPE attached to other Network Nodes, and a backbone layer that
   interconnects the NODES via a Digital Cross-Connect System that
   provide reconfigurable SONET circuits between the NODES (the use of
   circuits minizes delay and avoids the need for implementation of
   peak-transmission-rate packet switching).  Within this framework, the
   most likely places for near-term application of photonics, apart from
   pure transport (ie, 150 Mbps channels in a 2.4 Gbps SONET system),
   are in the Cross-Connect (a Wavelength Division Multiplexed based
   structure was described) and in next-generation LANs that provide
   Gigabit per second throughputs by use of multiple fibers, concurrent
   transmission, and new access mechanisms (such as store and forward).

   A planned interlocation Bell Labs multimedia gigabit/sec research
   network, LuckyNet, was described that attempts to extend many of the
   above concepts to achieve its principal goals: provision of a gigabit
   per second capability to a heterogeneous user community, the
   stimulation of applications that require Gpbs throughput (initial
   applications are video conferencing and LAN interconnect), and, to
   the extent possible, be based on standards so that interconnection
   with other Gigabit testbeds is possible.



Partridge                                                       [Page 6]

RFC 1152                  IRSG Workshop Report                April 1990


Session 4: High Speed Networks and the Phone System
           (David Tennenhouse, Chair)

   David Tennenhouse (MIT) reported on the ATM workshop he hosted the
   two days previous to this workshop.  His report will appear as part
   of the proceedings of his workshop.

   Wally St. John (LANL) followed with a presentation on the Los Alamos
   gigabit testbed.  This testbed is based on the High Performance
   Parallel Interface (HPPI) and on crossbar switch technology.  LANL
   has designed its own 16x16 crossbar switch and has also evaluated the
   Network Systems 8x8 crossbar switch. Future plans for the network
   include expansion to the CASA gigabit testbed.  The remote sites (San
   Diego Supercomputer Center, Caltech, and JPL) are configured
   similarly to the LANL testbed.  The long-haul interface is from HPPI
   to/from SONET (using ATM if in time).

   Wally also discussed some of the problems related to building a
   HPPI-SONET gateway:

      a)  Flow control.  The HPPI, by itself, is only readily extensible
          to 64 km because of the READY-type flow control used in the
          physical layer.  The gateway will need to incorporate larger
          buffers and independent flow control.

      b)  Error-rate expectations.  SONET is only specified to have a
          1E-10 BER on a per hop basis.  This is inadequate for long
          links.  Those in the know say that SONET will be much better
          but the designer is faced with the poor BER in the SONET spec.

      c)  Frame mapping.  There are several interesting issues to be
          considered in finding a good mapping from the HPPI packet
          to the SONET frame.  Some are what SONET STS levels will be
          available in what time frame, the availability of concatenated
          service, and the error rate issue.

   Dan Helman (UCSC) talked about work he has been doing with Darrell
   Long to examine the interconnection of Internet networks via an ATM
   B-ISDN network.  Since network interfaces and packet processing are
   the expensive parts of high-speed networks, they believe it doesn't
   make sense to use the ATM backbone only for transmission; it should
   be used for switching as well.  Therefore gateways (either shared by
   a subnet or integrated with fast hosts) are needed to encapsulate or
   convert conventional protocols to ATM format.  Gateways will be
   responsible for caching connections to recently accessed
   destinations.  Since many short-lived low-bandwidth connections as
   foreseen (e.g., for mail and ftp), routing in the ATM network (to set
   up connections) should not be complicated - a form of static routing



Partridge                                                       [Page 7]

RFC 1152                  IRSG Workshop Report                April 1990


   should be adequate.  Connection performance can be monitored by the
   gateways.  Connections are reestablished if unacceptable.  All
   decision making can be done by gateways and route servers at low
   packet rates, rather than the high aggregate rate of the ATM network.
   One complicated issue to be addressed is how to transparently
   introduce an ATM backbone alongside the existing Internet.

Session 5: Distributed Systems (David Farber, Chair)

   Craig Partridge (BBN Systems and Technologies) started this session
   by arguing that classic RPC does not scale well to gigabit-speed
   networks.  The gist of his argument was that machines are getting
   faster and faster, while the round-trip delay of networks is staying
   relatively constant because we cannot send faster than the speed of
   light.  As a result, the effective cost of doing a simple RPC,
   measured in instruction cycles spent waiting at the sending machine,
   will become extremely high (millions of instruction cycles spent
   waiting for the reply to an RPC).  Furthermore, the methods currently
   used to improve RPC performance, such as futures and parallel RPC, do
   not adequately solve this problem.  Future requests will have to be
   made much much earlier if they are to complete by the time they are
   needed.  Parallel RPC allows multiple threads, but doesn't solve the
   fact that each individual sequence of RPCs still takes a very long
   time.

   Craig went on to suggest that there are at least two possible ways
   out of the problem.  One approach is to try to do a lot of caching
   (to waste bandwidth to keep the CPU fed).  A limitation of this
   approach is that at some point the cache becomes so big that you have
   to keep in consistent with other systems' caches, and you suddenly
   find yourself doing synchronization RPCs to avoid doing normal RPCs
   (oops!).  A more promising approach is to try to consolidate RPCs
   being sent to the same machine into larger operations which can be
   sent as a single transaction, run on the remote machine, and the
   result returned.  (Craig noted that he is pursuing this approach in
   his doctoral dissertation at Harvard).

   Ken Schroder (BBN Systems and Technologies) gave a talk on the
   challenges of combining gigabit networks with wide-area heterogeneous
   distributed operating systems.  Ken feels the key goals of wide area
   distributed systems will be to support large volume data transfers
   between users of conferencing and similar applications, and to
   deliver information to a large number of end users sharing services
   such as satellite image databases.  These distributed systems will be
   motivated by the natural distribution of users, of information and of
   expensive special purpose computer resources.

   Ken pointed to three of the key problems that must be addressed at



Partridge                                                       [Page 8]

RFC 1152                  IRSG Workshop Report                April 1990


   the system level in these environments: how to provide high
   utilization; how to manage consistency and synchronization in the
   presence of concurrency and non-determinism; and how to construct
   scalable system and application services.  Utilization is key only to
   high performance applications, where current systems would be limited
   by the cost of factors such as repeatedly copying messages,
   converting data representations and switching between application and
   operating system.  Concurrency can be used improve performance, but
   is also likely to occur in many programs inadvertently because of
   distribution.  Techniques are required both to exploit concurrency
   when it is needed, and to limit it when non-determinism can lead to
   incorrect results.  Extensive research on ensuring consistency and
   resolving resource conflicts has been done in the database area,
   however distributed scheduling and the need for high availability
   despite partial system failures introduce special problems that
   require additional research.  Service scalability will be required to
   support customer needs as the size of the user community grow.  It
   will require attention both ensuring that components do not break
   when they are subdivided across additional processors to support a
   larger user population, and to ensure that performance does to each
   user can be affordably maintained as new users are added.

   In a bold presentation, Dave Cheriton (Stanford) made a sweeping
   argument that we are making a false dichotomy between distributed
   operating systems and networks.  In a gigabit world, he argued, the
   major resource in the system is the network, and in a normal
   operating system we would expect such a critical resource to be
   managed by the operating system.  Or, put another way, the gigabit
   network distributed operating system should manage the network.
   Cheriton went on to say that if a gigabit distributed operating
   system is managing the network, then it is perfectly reasonable to
   make the network very dumb (but fast) and put the system intelligence
   in the operating systems on the hosts that form the distributed
   system.

   In another talk on interprocess communication, Jonathan Smith (UPenn)
   again raised the problem of network delay limiting RPC performance.
   In contrast to Partridge's earlier talk, Smith argued that the
   appropriate approach is anticipation or caching.  He justified his
   argument with a simple cost example.  If a system is doing a page
   fetch between two systems which have a five millisecond round-trip
   network delay between them, the cost of fetching n pages is:

                         5 msec + (n-1) * 32 usec

   Thus the cost of fetching an additional page is only 32 usec, but
   underfetching and having to make another request to get a page you
   missed costs 5000 usec.  Based on these arguments, Smith suggested



Partridge                                                       [Page 9]

RFC 1152                  IRSG Workshop Report                April 1990


   that we re-examine work in virtual memory to see if there are
   comfortable ways to support distributed virtual memory with
   anticipation.

   In the third talk on RPC in the session, Tommy Joseph (Olivetti), for
   reasons similar to those of Partridge and Smith, argued that we have
   to get rid of RPC and give programmers alternative programming
   paradigms.  He sketched out ideas for asynchronous paradigms using