Minutes for RAID working group 11/11/93 Accredited Standards Committee X3, Information Processing Systems Doc. No.: X3T9.2/93-183 R0 Date: Nov 11, 1993 Project: Ref. Doc.: Reply to: G. Penokie To: Membership of X3T9.2 From: George Penokie and Ralph Weber Subject: Minutes of RAID Study Group Meeting (11/11/93) Agenda 1. Opening Remarks 2. Attendance and Membership 3. Approval of Agenda 4. Report on last RAID Working Group 5. Report on last RAID Advisor Board (RAB) meeting 6. SCSI-3 Addressing (93-161r0) 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 9:00 am, Thursday, 11 November 1993. He thanked John Lohmeyer and NCR 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 12th meeting of the 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. David Skinner O Advanced Micro Devices dave.skinner@amd.com Mr. Clifford Strang Bus Logic Mr. Michael Bryan Conner Peripherals Mr. Jim Whitworth Conner Peripherals Mr. Ed Haske CMD Mr. Bill Dallas Digital Equipment Corp. dallas@wasted.enet.dec.c om Mr. Doug Hagerman O Digital Equipment Corp. hagerman@starch.enet.dec. com Mr. Charles Monia P Digital Equipment Corp. Mr. Ralph Weber A Digital Equipment Corp. weber@star.enet.dec.com Mr. I. Dal Allen P ENDL 2501752@mcimail.com Mr. Ken Hallam A ENDL khallam@endlas.com Mr. Howard Grill O Formation Mr. Robert Liu Fujiitsu Mr. John Baudrexl Fujiitsu/Intellistor baudrexl@intellister.com Mr. Jeff Epstein Future Domain jeffd@fdc.mhs.compuserve .com Mr. Bill Hutchison O Hewlett Packard Co. hutch@.boi.hp.com Mr. Jeffrey Williams P Hewlett Packard Co. jlw@hpdmd48.boi.hp.com Mr. Paul Boulay O Hitachi Computer Products p_boulay@hitachi.com Mr. Sam Karunanithi Hitachi Micro Systems Mr. George Penokie P IBM Corp. gop@rchvmp3.vnet.ibm.com Mr. Giles Frazier O IBM Corp. gfrazier@asuvm6.vnet.ibm. com Mr. Charles Binford S NCR Corp. Charles.Binford@WichitaKS .NCR.COM Mr. Gene Milligan A Seagate Technology gene_milligan@notes.seaga te.com Mr. Gerald Houlder Seagate Technonogy Mr. Robert N. Snively P Sun Microsystems, Inc bob.snively@eng.sun.com Mr. Tak Asami Western Digital asami@dt.wdc.com 26 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. Report on last RAID Working Group Last RAID Working Group was held in West Palm Beach, FL. Several members of the RAID Advisory Board were present. Minutes of the Tarrytown meeting were published on the SCSI Reflector. 5. Report on last RAID Advisor Board (RAB) meeting RAID Advisory Board met the proceeding Tuesday (9 November). They reviewed the disk array model. George presented an action list produced at the RAB meeting. Two action items cover additional definitions needed in the model. George wants the definition process to complete by mid December. Additional action items were: -addressing (to be discussed today) -commands and mode pages -reported valid value ranges -error reporting -dual controller configurations (and reservations in such situations) -finalizing the general model (document X3T9.2/93-140) The overall goal is to take something to the March X3T9.2 (a.k.a. X3T10) Plenary meeting for some type of approval vote. 6. SCSI-3 Addressing (93-161r0) Penokie George presented a new document describing a RAID-centric SCSI-3 Addressing scheme (X3T9.2/93-161r0). The first lengthy discussion occurred when George called attention to the fact that LUN 0 becomes a non-data-transfer address. Discussion of the Address Translation Mode Page revealed the need for a null (no translation) entry in the LUN table. Concerns were raised about changing the LUN table while transfers are active. George proposed that changing the Active bit be prohibited while transfers are active. Changing a LUN table entry while transfers are active proved to be a stickier problem. Substituting for inactive LUN table entries seemed reasonable and George promised to make the right words for the next draft. George noted that the message code for Extended Identify code is 17h not BFh. This change resulted from a change in the conceptual design of Extended Identify. The Extended Identify does not allow disconnects to be disabled. A discussion of not allowing a second Expanded Identify produced a name change for the message. Expanded Identify became Expended LUN. Now, George turned to describing the disk array methodology for generating and interpreting the 64-bit LUNs in complex trees of disk arrays. The idea that succeeding lower layers strip-off and shift parts of the LUN confused several of those present. The actions performed both going down and up the tree was described. Dal commented that only those present at all of these meetings will be able to interpret (read) the final specification correctly. George next described the encoding of 8-byte disk array LUN. The 8-byte LUN is divided into 4 2-byte quantities. Thus, for layers can be represented in a single LUN value. The discussion quickly turned to how a single 2-byte might map to a LUN value in a traditional IDENTIFY message. This case is interest- ing because it offers an inexpensive way to build small capacity disk array controller systems. Topics discussed in this regard included: 1) cutting the LUN table to 32 entries and using the high-order LUN bit to select whether the table is used 2) whether the high-order bit selector is a 0 or a 1 to select using the table 3) whether both a 32 entry table and a 64 entry table must be supported by a single implementation (or, by the standard) 7. Action Items a) [Penokie] Review the SCSI-3 Addressing (93-161) document. 8. Meeting Schedule Starting in December 1993, the X3 and RAID Advisory Board meetings will be combined into a single meeting. The combined meeting will occur every month during the plenary week for either X3T9.2 or X3T9.3. The combined meeting will always be held on a Tuesday. Thus, the next two meetings will be at the Catamaran Hotel in San Diego, CA on 7 December 1993 and 11 January 1994. 9. Adjournment The meeting was adjourned at 11:44 pm. on Thursday, November 11 1993.