Accredited Standards Committee X3, Information Processing Systems Doc. No.: X3T9.2/93-108 R0 Date: June 23, 1993 Project: Ref. Doc.: Reply to: G. Penokie To: Membership of X3T9.2 From: George Penokie and Gerry Johnsen Subject: Minutes of RAID Study Group Meeting (6/23/93) Agenda 1. Opening Remarks 2. Attendance and Membership 3. Approval of Agenda 4. SCSI Disk Array Model (93-003r6) [Penokie] 6. RAID Issues from Trimm's Perspective (Gary Watson) [93-087r0] 7. Action Items 8. Meeting Schedule 9. Adjournment Results of Meeting 1. Opening Remarks George Penokie the RAID Study Group Chair, called the meeting to order at 8:00 am, Wednesday June 23, 1993. He thanked Joe Molina of Technology Forums for hosting the meeting. As is customary, the people attending introduced themselves. A copy of the attendance list was circulated for attendance and corrections. It was stated that the meeting had been authorized by X3T9.2 and would be conducted under the X3 rules. Ad hoc meetings take no final actions, but prepare recommendations for approval by the X3T9.2 task group. The voting rules for the meeting are those of the parent committee, X3T9.2. These rules are: one vote per company; and any participating company member may vote. The minutes of this meeting will be posted to the SCSI BBS and the SCSI Reflector and will be included in the next committee mailing. George stated that this is the seventh meeting of the RAID study group and the first joint meeting between the RAID Advisory Board Host Interface group and the X3T9.2 RAID study group. The purpose of the group is to deal with interface issues related to using RAIDs. The study group will assess the issues and then formulate a strategy for dealing with them. 2. Attendance and Membership Attendance at working group meetings does not count toward minimum attendance requirements for X3T9.2 membership. Working group meetings are open to any person or company to attend and to express their opinion on the subjects being discussed. The following people attended the meeting: RAID Study Group Meeting Attenders Name S Organization Electronic Mail Address ---------------------- - ------------------------- ------------------------- Mr. Randy Hall O Array Technology Corp. hall@arraytech.com Mr. Gerry Johnsen O Ciprico Inc. Mr. Bob Solomon O Data General Corp. bobsolomon@dgc.ceo.dg.com Mr. Paul Massiglia O Digital Equipment Corp. massiglia@genral.enet.dec .com Mr. Doug Hagerman O Digital Equipment Corp. hagerman@starch.enet.dec. com Mr. Don Johnson V Dynatek Mr. Howard Grill O Formation Mr. Jeffrey L. P Hewlett Packard Co. jlw@hpdmd48.boi.hp.com Williams Mr. Bill Hutchison O Hewlett Packard Co. hutch@.boi.hp.com Mr. David Snook O Hitachi America Mr. George Penokie P IBM Corp. gop@rchvmp3.vnet.ibm.com Mr. Giles Grazier O IBM Corp. gfrazier@asuvm6.vnet.ibm. com Mr. Keith Dulac O NCR Corp. keith.dulac@wichitaks.ncr .com Mr. Jerry Fredin O NCR Corp. jerry.fredin@wichitaks.nc r.com Mr. Dave Towle V Storage Dimensions davet@xstor.com Mr. Tom Ruwart V Unversity of MN/AHPCRC tmr@ahpcrc.umn.edu 16 people present Status Key: P - Principal A - Alternate O - Observer L - Liaison S,V - Visitor 3. Approval of Agenda The agenda developed at the meeting was approved. 4. SCSI Disk Array Model (93-003r6) [Penokie] George Penokie presented revision 6 of the SCSI Disk Array Model. This was the first time many on the people present had heard a presentation on the SCSI Disk Array Model and some time was spent mapping the terminology from the RAB host interface group to the SCSI Disk Array Model. Generally the terminology does map and the RAB group has agreed to convert their documentation to the SCSI Disk Array Model terminology in their next rev. There are a few concepts in the area of mapping LBAs to specific locations on specific devices which are not in the SCSI Model which George agreed should be in the SCSI Model. The RAB group will provide George with the necessary documentation and those new concepts will be placed into Rev 7 of the SCSI Model. A question was asked as to the difference between a DACL and a SDA. George could not think of any and will look into removing SDA from the model. There was a long discussion on how things should be addressed. The SCSI model uses an almost flat addressing scheme while the RAB model assumes an N deep layering scheme. Both groups agreed that neither scheme has been completely thought out. George agreed to exercise some scenarios to see if the flat scheme will work. Doug Hagerman indicated that DEC has a proposal which covers addressing which he will bring to the next meeting. There was a question as to weather or not all FRUs should be addressable. The answer is yes and no. All FRUs have to be identifiable so they can be called out for replacement but not all have to be addressable using LUIs. Multiple FRUs may be contained within a single LUI. It was decided that Number of LBAs need not be part of the maintenance services because that information is available through standard SCSI commands once the LUI of a device is known. The discussion on configuration services was the most controversial of the day. Opinions on how configuration should be handled ranged from; 'should all be vendor specific' to 'should be controlled down to the sector using bit maps' to 'should be characteristics passed to the DACL with the DACL figuring out the details.' It was generally agreed that: -some definition is needed to allow interoperability -An expert system concept (like the SCSI model) is a good long term concept but is not practical today -Using RAID terminology is not desirable -Defining the RAID concepts (0 through 5) without using 'RAID' and allowing for future concepts is desirable The conclusion was that a limited expert system might me made to work but there would have to be an alternate path for detailed mapping control. It is safe to say not everyone agreed with this conclusion. Another comment about the configuration services was that the model should preclude 'on the fly' changes in the configuration. It was agreed to add to the SCSI Disk Array Services the following: -Verify LBA Set -Enable LBA Range -Enable Writes to a LBA Range The reporting of exception conditions was discussed and it was verified that this indeed is a religious issues and can only be solved by allowing the host to select which religion it wants to follow. Rev. 6 of the SCSI Disk Array model already covers this issue and no change is requested in this area. The remaining time was used to answer questions about sections 6 and 7 of the model. A few minor wording changes where agreed to. 5. RAID Issues from Trimm's Perspective (Gary Watson) [93-087r0] Gary Watson was not present. No discussion of 93-087r0 occurred. 6. Action Items 1) George Penokie will prepare revision 7 of the RAID Model. 7. Meeting Schedule The next meeting of the RAID Study Group is planned for July 22, 1993 at the Sheraton Tara Wayfarer Inn in Bedford, NH. The meeting is expected to start at 9:00am-12:00am. George stated that in the even numbered months the RAID Study Group will plan to meet with the RAID Advisory Board. In the odd numbered months the RAID Study Group will meet on Thursday morning with X3T9.2. 8. Adjournment The meeting was adjourned at 11:00 a.m. on Wednesday June 23, 1993.