bibliography

[Rod Van Meter <rdv@Network-Alchemy.COM>]

Okay, here are some bibliographic entries for some useful papers on
transport protocols and other topics.  For reasons probably related to
my job change but hopefully not meaning I've lost a bunch of it, my
current bibliography appears incomplete.  I know I'm missing a survey
of lightweight transport protocols (from IEEE ToN, I think) and a few
others (my total bibliography is over 400 papers, but for some reason,
I'm always missing whatever today's most important one is :-().

I'll immodestly start with the non-transport papers:

All of the Netstation papers are available at
http://www.isi.edu/netstation/

@InProceedings{van-meter:visa,
  author = 	 "Rodney Van{ }Meter and Greg Finn and Steve Hotz",
  title = 	 "{VISA}: Netstation's Virtual Internet {SCSI} Adapter",
  crossref =	 "asplos98"
}

@InProceedings{van-meter:ip-for-naps,
  author = 	 "Rodney Van{ }Meter and Greg Finn and Steve Hotz",
  title = 	 "Internet Protocols for Network-Attached Peripherals",
  crossref = "kobler:gsfc-msst-98"
}

A one-pager on why IP:

@InProceedings{van-meter:ip-naps,
  author =	 "Rodney Van{ }Meter and Steve Hotz and Gregory G. Finn",
  title = 	 "Task Force on Network Storage Architecture:
		  Internet-attached storage devices",
  crossref =	 "hicss-30",
  note =	 "White paper for the task force that met in
		  conjunction with the conference",
  URL =		 {http://www.isi.edu/netstation/hicss_isi.ftp.ps}
}

This one is now quite dated:

@ARTICLE{van-meter:nap-overview,
title = {A Brief Survey of Current Work on Network Attached
Peripherals (extended abstract)},
author = {Rodney Van{ }Meter},
journal = {ACM Operating Systems Review},
year = 1996,
month = jan,
pages = {63--70}
}

Garth's stuff (these are the two best, IMHO):

@TechReport{gibson:nasd-case,
  author = 	 {Garth Gibson and others},
  title = 	 {A Case for Network-Attached Secure Disks},
  institution =  {CMU},
  year = 	 1996,
  number = 	 {CMU-CS-96-142},
  month = 	 jun,
  comment = 	 {Good paper presenting a taxonomy of NAP interfaces,
                  and numbers evaluating potential performance
                  improvements for NFS and AFS, varing from a few
                  percent to a factor of ten in CPU cycles.},
  read = 	 {1996/7/24}
}

@InProceedings{gibson:nasd-file-server-scaling,
  author = 	 "Garth A. Gibson and others",
  title = 	 "File Server Scaling with Network-Attached Secure Disks",
  crossref =	 "sigmetrics97",
  comment =	 {Presents their NASD taxonomy again, with more
		  detailed analysis of the results of using their NASD 
		  model.  Suggests total server workload reduction of
		  30 percent or so for the ``NetSCSI'' approach (which 
		  they don't recommend) and factors of 5 and 10 
		  for AFS and NFS using their particular NASD model.
		  Replays UCB NFS traces and their own AFS traces
		  against an emulated server to test response time.
		  Disparages UDP performance.  Great references.},
  read =	 {1997/10/26},
  URL =		 {http://www.cs.cmu.edu/Web/Groups/NASD/},
  location =	 { rdv : folder : naps }
}


Okay, now transport protocols:


@Article{baguette92:_xtp_conn_compare,
  author = 	 {Yves Baguette and Andre Danthine},
  title = 	 {Comparison of {TP4}, {TCP} and {XTP} -- Part 1:
                  Connection Management Mechanisms},
  journal = 	 {European Transactions on Telecommunications},
  year = 	 1992,
  volume =	 3,
  number =	 5,
  month =	 {Sept.-Oct.},
  comment =	 {English is a little rough, but good info.
                  Concentrates mostly on TP4 v. XTP.  TP4, like TCP,
                  uses a 3-way handshake.  XTP requires an explicit
                  connection open packet (different format than data
                  packets), but data packets can be pipelined with
                  it, cutting out a round trip.  Connection close can
                  also be reduced to 2-way handshake.},
  location =	 { rdv : folder : XTP },
  read =	 {1998/4/13},
  URL =		 {http://www-run.montefiore.ulg.ac.be/publications/papers/abstract-R92-03_A.html}
}

@Article{baguette92:_xtp_data_compare,
  author = 	 {Yves Baguette and Andre Danthine},
  title = 	 {Comparison of {TP4}, {TCP} and {XTP} -- Part 2:
                  Data Transfer Mechanisms},
  journal = 	 {European Transactions on Telecommunications},
  year = 	 1992,
  volume =	 3,
  number =	 5,
  month =	 {Sept.-Oct.},
  comment =	 {Spends a lot of time on the differences between XTP
                  3.5 and 3.6 (current version is 4.0).  Provides some
                  support for ALF, in the form of an end-of-message
                  bit.  No segmentation/reassembly allowed in the
                  net.  Checksums are in trailers, and are optional.
                  Uses a timeout to decide a packet has been lost, but
                  the details of that setting aren't covered.  Error
                  handling is tricky in standard internets because it
                  assumes in-order delivery (which requires XTP-aware
                  routers under some conditions).  Uses a window
                  mechanism for flow control.  Supports optional
                  FASTNACKs (again depends on in-order delivery).
                  ACKs are piggybacked.  Congestion control is done
                  via path rate control, which again requires
                  XTP-aware routers.  XTP supports both reliable and
                  unreliable multicast.},
  location =	 { rdv : folder : XTP },
  read =	 {1998/4/13},
  URL =		 {http://www-run.montefiore.ulg.ac.be/publications/papers/abstract-R92-03_B.html}
}

@INPROCEEDINGS{cheriton:vmtp,
author = {Cheriton, David R.},
title = {Exploiting Recursion to Simplify RPC Communication
Architectures},
booktitle = {Proc. ACM SigComm (Stanford, CA August 1988)},
year = 1988,
month = aug,
pages = {76--87}}

@MISC{cheriton:vmtp-rfc,
author = {Cheriton, David R.},
title = {VMTP: Versatile Message Transaction Protocol / Protocol
Specification},
number = {Internet RFC 1045},
organization = {USC/ISI},
year = 1988,
month = feb}

@Manual{hippi-st,
  title = 	 {Information Technology - Scheduled Transfer Protocol
                  {(ST)} {T11.1/Project 1245-D}},
  organization = {NCITS},
  year =	 1998,
  month =	 jan,
  note =	 {rev 1.45},
  comment =	 {Scheduled transfers for HIPPI, as a transport
                  protocol.  Also claimed to work for FC and ethernet,
                  but I dpn't think anybody's really using that yet.
                  Separate control and data connections.  Begins with
                  a setup negotiation, in which the destination
                  indicates how much data it's prepared to receive.
                  Mentions that this could be adjusted downward by
                  intermediate nodes, but no discussion of how.
                  Assumes very reliable in-order delivery for
                  efficiency, though should work on packet loss due to
                  ACKs (apparently, one ACK per data pkt is required).
                  Targetted to minimize the work on receive.
                  Transfers consist of blocks which consist of STUs,
                  which I think correspond to packets.  Must be a
                  power of two in size.  Some support for striping,
                  including fan-in and fan-out, but still looks incomplete.
                  Headers indicate when the upper-layer system should
                  be interrupted.  Some support for negotiated
                  out-of-order delivery, but I'm not sure how that
                  works with the max received index in the ACK packets
                  (ACK pkts are primarily indicators of buffer
                  availability for flow control, really CTS).  Says it
                  assumes low latency.  No discussion of how packets
                  might overlap in transmission; diagrams show a
                  series of data pkts then a series of CTSes.  Thinks
                  it ld be used to carry TCP/IP traffic, but I view it
                  more as a transport protocol and would be curious to
                  see ST over IP.  Despite not being an Internet
                  protocol, bows to the IANA for selection of things
                  like well-known port numbers for services.},
  location =	 { rdv : folder : hippi },
  read =	 {1998/1/30},
  URL =          {http://www.cic-5.lanl.gov/~det/dST145.pdf}
}

There are a number of other papers on XTP, as well as the spec,
available at http://www.ca.sandia.gov/xtp/.

Hope this helps.