Subject: Re: Scheduling solutions?
From: Lindsay Moore <>
Date: Thu, 22 Jan 2004 18:22:55 -0800

I agree that it would be very useful to have a much shorter max time per 
week for each user such as 8 or 10hrs.  Clearly, as with other equipment at 
SNF (i am thinking specifically of the KarlSuss mask aligner) there can be 
exceptions in special cases.  It is my impression that many people use the 
Raith gratuitously, doing very long and involved exposure matrices or not 
getting help from more experienced users, leading to a guess-and-check 
writing scheme in which many writes are done when one could have been done 
if more experienced users had helped out.  I think that a more restrictive 
time limit would result in people using the Hitachi more often (when it is 
working and if their write allows), people loading their samples in groups 
(thereby saving 1/2 hr) and considering more carefully how long they really 
need for their write.  Yup.


At 02:08 PM 1/22/2004 -0800, John Shott wrote:
>John, Hendrik, and all Raith Users:
>I certainly welcome discussions related to improved ways of reserving the 
>Raith.  As I'm sure many of you are painfully aware, there are a handful 
>of tools in the lab for which reservation "allocation" becomes a problem.
>My own preference would be to try to avoid solutions that avoid staff time 
>to administer and I would hope that we could come up with a set of 
>reservation rules that could be enforced or checked in an automated way.
>I want to make you aware that Bill Murray is currently working on putting 
>the infrastructure in place that will allow us to specify and check 
>equipment-specific reservation rules.  While it will likely take several 
>months before something is ready for production, I can also tell you that 
>this capability has been on our wish list for some time and has finally 
>gotten to the top of the list.
>I would encourage you to continue to think of rules or collections of 
>rules that you think might offer the means of allowing fair allocation of 
>scarce equipment time.  I expect that we will be able to accommodate 
>AND/OR combinations of different rules.  In many respects easily expressed 
>rules may offer a good place to start ... easily expressed probably 
>translates into easily understood and easily implemented.  Examples of 
>easily expressed rules might include:
>1. A limit of X total hours of reservation time during the coming week.
>2. Maximum individual reservation duration of 4 hours during "prime time" 
>and 8 hours during other times. (For some definition of prime time such as 
>M-F, 9am-6pm).
>3. Cancellation of a reservation less than X hours in advance penalizes 
>the ability to make reservations for a time into the future.
>Some of the things that may be harder to do would include some sort of 
>waitng list or priority-based allocation ... at the moment we have no 
>support for "bidding" for reservations or a waiting list.  While that 
>could be added, if need by, my guess is that we can improve many things, 
>from a reservation standpoint, by being able to define and enforce 
>collections of simpler rules.
>In any event, I wanted you all to know that support for improved means of 
>allocating reservations is important to Team Coral and we are beginning to 
>work on tools that will, we believe, have a positive impact on many if not 
>all of the heavily used tools such as the Raith, STS etcher, LPCVD 
>nitride, etc.