[comp.os.research] Distributed time.

page@CS.UCLA.EDU (Thomas Page) (01/15/91)

I am co-teaching a graduate seminar on distributed systems.  I would like
to encorporate a lecture on the notion of time in a distributed system, but
I am woefully ignorant on the subject.  I would appreciate any and all
suggestions on what you thing should be covered, and especially any good
references.  At this point I intend to cover Jefferson's Virtual Time, and
protocols for maintaining clock synchronization (which ones?).

	Many thanks,
		Tom Page
 

jward@SRC.Honeywell.COM (Jon Ward) (01/17/91)

There are a number of classic papers on time and distributed systems
written by Leslie Lamport.  He has also co-authored several papers with
Mani Chandy.  The following is a favorite introductory paper on the topic
and should be required of any graduate course on the topic:

Lamport, L., "Time, Clocks, and the Ordering of Events in a Distributed
System," CACM, July 1978, Vol. 21, No. 7.

vera%chook.adelaide.edu.au@augean.ua.OZ.AU (Vera M) (01/23/91)

In article <11132@darkstar.ucsc.edu>, page@CS.UCLA.EDU (Thomas Page) writes:
|> 
|> 
|> I am co-teaching a graduate seminar on distributed systems.  I would like
|> to encorporate a lecture on the notion of time in a distributed system, but
|> I am woefully ignorant on the subject.  I would appreciate any and all
|> suggestions on what you thing should be covered, and especially any good
|> references.  At this point I intend to cover Jefferson's Virtual Time, and
|> protocols for maintaining clock synchronization (which ones?).
|> 
|> 	Many thanks,
|> 		Tom Page
|>  

I found the follwoing rather suitable:

L.Lamport
Time, clocks and the Ordering of Events            
Distributed Systems.                              
Communications of the ACM, July 1978.                        

H. Kopetz and W. Ochsenreiter
Clock synchronization in Distributed Real-Time Computer Systems
IEEE Transactions on Computers, VolC-36, No8, August 1987.

P. Ramanathan, D.D. kandlur, and K.G. Shin
Hardware-Assisted Software Clock Synchronization 
for Homogeneous distributed Systems
IEEE Transactions on Computers, Vol 39, No4, April 1990.

Hope that is of help.

Vera

kim@bilby.cs.uwa.oz.au (Kim Shearer) (01/24/91)

In <11456@darkstar.ucsc.edu> vera%chook.adelaide.edu.au@augean.ua.OZ.AU (Vera M) writes:


>In article <11132@darkstar.ucsc.edu>, page@CS.UCLA.EDU (Thomas Page) writes:
>|> 
>|> 
>|> I am co-teaching a graduate seminar on distributed systems.  I would like
>|> to encorporate a lecture on the notion of time in a distributed system, but
>|> I am woefully ignorant on the subject.  I would appreciate any and all
>|> suggestions on what you thing should be covered, and especially any good
>|> references.  At this point I intend to cover Jefferson's Virtual Time, and
>|> protocols for maintaining clock synchronization (which ones?).
>|> 
>|> 	Many thanks,
>|> 		Tom Page
>|>  

 Im not sure whether you will find this suitable but maybe you
 could include some of the work on causal ordering.

 You may be including this anyway, but it seems to me to be an
 attempt to simplify distributed time and leave only the useful
 concepts.

 Some references you might like to look at :

   Exploiting Replication
	       K P Birman  T A Joseph
	       Not sure how to get your hands on it, try Ken Birman 
	       at Cornell. This is dated June 1 1988

   The use of Efficient Broadcast Protocols in Asynchronous 
   Distributed Systems
	       F B Schmuck
	       PhD Thesis TR 99-928

ken@cs.cornell.edu (Ken Birman) (01/25/91)

In article <11525@darkstar.ucsc.edu> kim@bilby.cs.uwa.oz.au (Kim Shearer) writes:
> ....
> Some references you might like to look at :
>
>   Exploiting Replication
>	       K P Birman  T A Joseph
>	       Not sure how to get your hands on it, try Ken Birman 
>	       at Cornell. This is dated June 1 1988
>
>   The use of Efficient Broadcast Protocols in Asynchronous 
>   Distributed Systems
>	       F B Schmuck
>	       PhD Thesis TR 99-928

Copies of these are available from Cornell.  Email to isis@cs.cornell.edu
and include a postal address.  Except for the ISIS manual, all Cornell TR's
are provided without charge.

The Exploiting Replication paper actually appeared in the textbook that
Sape Mullender edited for Addison Wesley: Distributed Systems (ACM Press).

We have a number of other papers on this and related issues; the full list
is available from the isis secretary.  New papers are announced by 
postings to comp.sys.isis and can be copied using anonymous FTP from
our repository (anonymous FTP from cu-arpa.cs.cornell.edu, pub directory).
These old papers, however, are only available in hard copy.

If you get interested in systems that enforce casual ordering, you'll
want to read some other papers too:
	Psync (Larry Peterson et. al., Arizona)
	Lazy Replication (Rivka Ladin et. al., MIT/DEC CRL)
	John Hennesey's work on lazy memory (Stanford; hardware)
	Causal Memory (Hutto et. al -- Georgia Tech)
	Fast causal multicast (recent Cornell TR)
	Delta-4 (Esprit multicast architecture)
	ANSA (Esprit Advanced Network Systems Architecture)
I'm probably leaving out some obvious references, but this is a start...