Transferring scheduling data from a plurality of disk storage devices to a network switch before transferring data associated with scheduled requests between the network switch and a plurality of host initiators

1. A network switch for resolving requests from a plurality of host initiators by scheduling access to a plurality of disk storage devices, the network switch comprising:

(a) a switched fabric comprising a plurality of switching elements, each switching element comprising:

a plurality of bi-directional switched fabric ports; and

a control input connected to receive switch control data for selectively configuring the switching element in order to interconnect the bi-directional switched fabric ports;

(b) a memory for storing a routing and scheduling program; and

(c) a microprocessor, responsive to the requests, for executing the steps of the routing and scheduling program to generate the switch control data to transmit scheduled requests through the bi-directional switched fabric ports, wherein:

at least one of the plurality of switching elements comprises a disk storage interface for connecting to a selected one of the disk storage devices;

the microprocessor for scheduling access to the plurality of disk storage devices through the disk storage interface;

the disk storage interface for receiving scheduling data from the selected one of the disk storage devices;

the memory for receiving the scheduling data via the bi-directional switched fabric ports of a selected number of the switching elements;

the scheduling data is processed according to a priority such that the selected switching elements transfer the scheduling data through the bi-directional switched fabric ports before transferring data associated with the scheduled requests;

each disk storage device comprises a disk and a head; and

the scheduling data comprises a radial location of the head relative to the disk within each disk storage device.

2. The network switch as recited in claim 1 , wherein the at least one switching element further comprises a disk storage device connected to the disk storage interface.

3. The network switch as recited in claim 1 , wherein the scheduling data further comprises a circumferential location of the head relative to the disk within each disk drive.

4. The network switch as recited in claim 1 , wherein the switching elements further comprise a plurality of virtual lanes, wherein:

(a) at least one of the virtual lanes is reserved for transferring data associated with the scheduled requests;

(b) at least one of the virtual lanes is reserved for transferring the scheduling data; and

(c) the virtual lane for transferring the scheduling data comprises a higher priority than the virtual lane for transferring the data associated with the scheduled requests.

5. The network switch as recited in claim 1 , wherein the scheduling data is communicated to the memory through the bi-directional switched fabric ports according to an isochronous protocol.

6. A method of resolving requests from a plurality of host initiators by scheduling access to a plurality of disk storage devices connected to a network switch, the network switch comprising a switched fabric comprising a plurality of switching elements, the method comprising the steps of:

(a) transmitting through the switching elements scheduling data from the plurality of disk storage devices to a memory;

(b) evaluating the scheduling data in order to schedule the requests from the host initiators; and

(c) transmitting data associated with the scheduled requests through the switching elements to the plurality of disk storage devices, wherein:

the scheduling data is processed according to a priority such that the switching elements transfer the scheduling data before transferring data associated with the scheduled requests;

each disk storage device comprises a disk and a head; and

the scheduling data comprises a radial location of the head relative to the disk within each disk storage device.

7. The method as recited in claim 6 , wherein the scheduling data further comprises a circumferential location of the head relative to the disk within each disk drive.

8. The method as recited in claim 6 , wherein the switching elements further comprise a plurality of virtual lanes, wherein:

(a) at least one of the virtual lanes is reserved for transferring data associated with the scheduled requests;

(b) at least one of the virtual lanes is reserved for transferring the scheduling data; and

(c) the virtual lane for transferring the scheduling data comprises a higher priority than the virtual lane for transferring the data associated with the scheduled requests.

9. The method as recited in claim 6 , wherein the scheduling data is communicated to the memory through the switching elements according to an isochronous protocol.