Method and apparatus for implementing high-performance, scaleable data processing and storage systems

• US 8,825,882 B2
• Filed: 03/20/2006
• Issued: 09/02/2014
• Est. Priority Date: 06/30/2000
• Status: Expired due to Fees

• Current Assignee: Emc IP Holding Company LLC (Dell Technologies Inc.)
• Original Assignee: EMC Corporation (Dell Technologies Inc.)
• Inventors: Bopardikar, Raju C., Bast, Jacob Y., Cardone, Gary A., Kaufman, David E., MacEachern, Stuart P., McLeod, Bruce D., Nolan, James M., Radouch, Zdenek, Stiffler, Jack J., Wentworth, James A.
• Primary Examiner(s): MIRZA, ADNAN M
• Application Number:
  US11/384,756
• Publication Number:
  US 20060161642A1
• Time in Patent Office: 3,088 Days
• Field of Search: 709/103, 709/102, 710/1
• US Class Current: 709/229
• CPC Class Codes: G06F 2206/1012 Load balancing
• CPC Class Codes: G06F 3/061 Improving I/O performance
• CPC Class Codes: G06F 3/0635 by changing the path, e.g. traffic rerouting, path reconfiguration
• CPC Class Codes: G06F 3/067 Distributed or networked storage systems, e.g. storage area networks [SAN], network attached storage [NAS]
• CPC Class Codes: G06F 9/505 considering the load
• CPC Class Codes: G06F 9/5055 considering software capabilities, i.e. software resources associated or available to the machine
• CPC Class Codes: H04L 12/54 Store-and-forward switching systems packet switching systems H04L45/00, H04L47/00
• CPC Class Codes: H04L 41/0695 the faulty arrangement being the maintenance, administration or management system

Abstract

A data system architecture is described that allows multiple processing and storage resources to be connected to multiple clients so as 1) to distribute the clients'"'"' workload efficiently across the available resources; and 2) to enable scaleable expansion, both in terms of the number of clients and in the number of resources. The major features of the architecture are separate, modular, client and resource elements that can be added independently, a high-performance cross-bar data switch interconnecting these various elements, separate serial communication paths for controlling the cross-bar switch settings, separate communication paths for passing control information among the various elements and a resource utilization methodology that enables clients to distribute processing or storage tasks across all available resources, thereby eliminating “hot spots” resulting from uneven utilization of those resources.

Claims

1. A fault-tolerant computer system for providing scalable data processing and storage services to a client from a plurality of storage resources, comprising:

a plurality of identical resources interface modules connected to the storage resources;

a plurality of identical access interface modules which receive requests for service from the client and select a subset of the plurality of resource interface modules to provide the requested service and distribute the workload across the plurality of storage resources; and

a switch for temporarily connecting one of the plurality of access interface modules to the selected subset of the plurality of resource interface modules for providing the service to the client, the switch being constructed in two identical halves, each switch half is enabled to be connected to each of the resource interface modules and to each of the access interface modules so that each switch half supports a full data bandwidth between all pairs of modules, the two switch halves being interconnected so that a failure in one switch does not make the computer system inoperative.

2. The computer system of claim 1, wherein one switch half is designated as active and the other switch half is designated as standby and the active switch half is used to temporarily connect one of the plurality of access interface modules to the selected subset of the plurality of resource interface modules.

3. The computer system of claim 2, wherein the active switch half and the standby switch half exchange roles if the active switch half fails.

4. The computer system of claim 1, wherein each storage resource is connected to at least two resources interface modules so that a failure in any resource module does not prevent access to the each storage resource.

5. The computer system of claim 1, wherein each access interface module can assume the workload of any of the plurality of access interface modules so that a failure in any access interface module can be bypassed by assigning the workload of the failed module to another of the access interface modules.

6. The computer system of claim 1, wherein each resource interface module can assume the workload of any of the plurality of resource interface modules so that a failure in any resource interface module can be bypassed by assigning the workload of the failed module to another of the resource interface modules.

7. The computer system of claim 1, wherein data is stored on one of the plurality of storage resources and the computer system further comprises a parity generator which computes a data tag, including parity information, from the data and stores the data tag on the plurality of storage resource apart from the data.

8. The computer system of claim 1, wherein data is stored on one of the plurality of storage resources and an acknowledgement is returned to the client after the data has been stored and wherein the data is stored on at least two separate storage resources before the acknowledgement is returned to the client.

9. A method for providing fault-tolerant scalable data processing and storage services to a client from a plurality of storage resources, comprising:

(a) providing a plurality of identical resource interface modules connected to the storage resources;

(b) providing a plurality of identical access interface modules which receive requests for service from the client and select a subset of the plurality of resource interface modules to provide the requested service and distribute the workload across the plurality of storage resources; and

(c) using a switch for temporarily connecting one of the plurality of access interface modules to the selected subset of the plurality of resource interface modules for providing the service to the client, the switch being constructed in two identical halves, each switch half is enabled to be connected to each of the resource interface modules and to each of the access interface modules so that each switch half supports a full data bandwidth between all pairs of modules, the two switch halves being interconnected so that a failure in one switch does not make the computer system inoperative.

10. The method of claim 9, where step (c) comprises designating one switch half as active and designating the other switch half as standby and using the active switch half to temporarily connect one of the plurality of access interface modules to the selected subset of the plurality of resource interface modules.

11. The method of claim 9, further comprising:

(d) exchanging the roles of the active switch half and the standby switch half if the active switch half fails.

12. The method of claim 9, wherein step (a) comprises connecting each storage resource to at least two resource interface modules so that a failure in any resource interface module does not prevent access to the each storage resource.

13. The method of claim 9, wherein step (b) comprises providing access interface modules that can assume the workload of any other of the plurality of access interface modules so that a failure in any access interface module can be bypassed by assigning the workload of the failed module to another of the access interface modules.

14. The method of claim 9, wherein step (a) comprises providing resource interface modules that can assume the workload of any other of the plurality of resource interface modules so that a failure in any resource interface module can be bypassed by assigning the workload of the failed module to another of the resource interface modules.

15. The method of claim 9, wherein data is stored on one of the plurality of storage resources and the method further comprises:

(e) computing a data tag, including parity information, from the data; and

(f) storing the data tag on the plurality of storage resource apart from the data.

16. The method of claim 9, wherein data is stored on one of the plurality of storage resources and an acknowledgement is returned to the client after the data has been stored and wherein the method further comprises:

(g) storing data on at least two separate storage resources before the acknowledgement is returned to the client.