Storage system

• US 6,446,162 B1
• Filed: 10/28/1999
• Issued: 09/03/2002
• Est. Priority Date: 10/30/1998
• Status: Expired due to Term

• Current Assignee: Gula Consulting Limited Liability Company (Intellectual Ventures LLC)
• Original Assignee: Koninklijke Philips N.V.
• Inventors: Coumans, Pascal F. A., Korst, Johannes H. M.
• Primary Examiner(s): Yoo, Do Hyun
• Assistant Examiner(s): GOODARZI, NASSER MOAZZAMI
• Application Number:
  US09/428,774
• Time in Patent Office: 1,041 Days
• Field of Search: 711/114, 710/6, 369/34, 365/200
• US Class Current: 711/114
• CPC Class Codes: G06F 11/1076 Parity data used in redundant arrays of independent storages, e.g. in RAID systems
• CPC Class Codes: G06F 11/1088 Reconstruction on already foreseen single or plurality of spare disks
• CPC Class Codes: G06F 3/061 Improving I/O performance
• CPC Class Codes: G06F 3/064 Management of blocks
• CPC Class Codes: G06F 3/0689 Disk arrays, e.g. RAID, JBOD
• CPC Class Codes: H04N 21/2182 involving memory arrays, e.g. RAID disk arrays RAID arrays per se G06F3/0689; use of parity to protect data in RAID systems G06F11/1008
• CPC Class Codes: H04N 21/2318 using striping
• CPC Class Codes: H04N 21/2326 Scheduling disk or memory reading operations

Abstract

The invention relates to a storage system (100) such as a video on demand system, comprising a plurality of storage units (110-120) such as disk drives. Each one of the storage units (110-120) has a number of storage zones (121-126) with mutually different expected data retrieval times. Data is stored in data units, with each data unit comprising N blocks, wherein N≧2. The blocks of a data unit are distributed over the storage units (110-120). The blocks comprise redundant information such that any one of a plurality of selections comprising N−K of the N blocks suffices for retrieving the data unit, wherein K≧1. For retrieving a data unit from the storage units (110-120), a reader determines which N−K of the N blocks to retrieve on the basis of a selection procedure. In accordance with the invention, the selection procedure takes into account the expected data retrieval times in order to optimize bandwidth utilization.

Claims

1. Storage system (100) comprising:

a plurality of storage units (110-120), each one of which comprising a plurality of storage zones (121-126) having mutually different expected data retrieval times, at least one data unit being stored in said plurality of storage units (110-120), said data unit comprising N blocks, wherein N≧

2, the blocks comprising redundant information such that any one of a plurality of selections comprising N−

K of the N blocks suffices for retrieving the data unit, wherein K≧

1; and

a reader for retrieving said data unit from the plurality of storage units (110-120), the reader being arranged for determining a particular one of said plurality of selections on the basis of a selection procedure, the selection procedure taking into account said expected data retrieval times.

2. Storage system as claimed in claim 1, wherein the N blocks are stored in N different ones of said storage units (110-120) and the selection procedure further takes into account a momentary load distribution in said plurality of storage units (110-120).

3. Storage system as claimed in claim 2, wherein the selection procedure in a first step aims at load balancing and in a second step aims at improvement of bandwidth utilization.

4. Storage system as claimed in claim 2, the selection procedure optimizing in which both the momentary load distribution of the system and the expected retrieval times are accounted for.

5. A storage system (100) comprising:

a plurality of storage units (110-120), each one of which comprising a plurality of storage zones (121-126) having mutually different expected data retrieval times;

a loader for storing a data unit in said plurality of storage units, said loader being arranged for converting said data unit into N blocks, wherein N≧

2, the blocks comprising redundant information such that any one of a plurality of selections comprising K of the N blocks suffices for retrieving the data unit, wherein K≧

1 and for storing said N blocks in said plurality of storage units such that each storage unit comprises at most one of said N blocks and that said N blocks are distributed at least approximately equally among faster and slower ones of said storage zones.

6. A method of retrieving data units from a system (100) comprising a plurality of storage units (110-120), each one of which comprising a plurality of storage zones (121-126) having mutually different expected data retrieval times, at least one data unit being stored in said plurality of storage units (110-120), said data unit comprising N blocks, wherein N≧

2, the blocks comprising redundant information such that any one of a plurality of selections comprising N−

K of the N blocks suffices for retrieving the data unit, wherein K≧

1, the method comprising the steps of;

determining a particular one of said plurality of selections, thereby taking into account said expected data retrieval times; and

retrieving the blocks pertaining to that particular selection.

7. Method of storing a data unit in a system (100) comprising a plurality of a storage units (110-120), each one of which comprising a plurality of storage zones (121-126) having mutually different expected data retrieval times, the method comprising the steps of:

converting said data unit into N blocks, wherein N≧

2, the blocks comprising redundant information such that any one of a plurality of selections comprising N−

K of the N blocks suffices for retrieving the data unit, wherein K≧

1;

storing said N blocks in said plurality of storage units such that each storage unit comprises at most one of said N blocks and that said N blocks are distributed at least approximately equally among faster and slower ones of said storage zones.