Accelerator system for use with secure data storage
View Patent ↗Data processing and an accelerator system therefore are described. An embodiment relates generally to a data processing system. In such an embodiment, a bus and an accelerator are coupled to one another. The accelerator has an application function block. The application function block is to process data to provide processed data to storage. A network interface is coupled to obtain the processed data from the storage for transmission.
1. A computer system, comprising:
a general-purpose processor for execution of a user application in an application mode and kernel-mode drivers in a kernel-mode, wherein the kernel-mode drivers include a class driver and a filter driver; and
an accelerator system coupled to the general-purpose processor via a first bus;
wherein the class driver is in communication with the user application to receive a request packet to provide a request block in response to the request packet;
wherein the filter driver is in communication with the class driver to receive the request block;
wherein the request block includes a system payload pointer that points to a first location of a data set stored in system memory and a write command; and
wherein for the write command a configured Programmable Logic Device of the accelerator system is to:
retrieve the data set from the system memory using the system payload pointer;
generate at least two portions of data from the data set;
store the at least two portions of data in local memory of the accelerator and;
transmit the at least two portions of data for storage in a cloud computing storage network having at least two remote storage devices; wherein the data set is restorable from at least a plurality of the at least two portions of data by recombining data from the plurality of the at least two portions of data.
2. The system of claim 1 , wherein:
the kernel-mode drivers include a device driver, a stack driver, and a port driver; and
the filter driver generates first commands and second commands, each of the first commands and the second commands include a same set of local payload pointers generated by the filter driver;
the device driver is in communication with the filter driver to receive the system payload pointer and the first commands; and
the device driver is in communication with the accelerator system to provide the first commands and the payload pointer to the accelerator system.
3. The system of claim 2 , wherein the at least two portions are stored at locations in the local memory of the accelerator pointed to by the local payload pointers of the first commands.
4. The system of claim 2 , wherein:
the stack driver is in communication with the filter driver to receive the second commands;
the port driver is in communication with the stack driver to receive the second commands; and
the accelerator system is in communication with the port driver to receive the second commands.
5. The system of claim 2 , wherein:
the port driver is in communication with a network interface of the accelerator system;
the network interface is coupled to the local memory using the local payload pointers of the second commands to transmit the at least two portions of data for storage in storage devices coupled to the cloud computing storage network.
6. The system of claim 5 , wherein the cloud computing storage network includes at least two respectively remote storage devices.
7. The system of claim 5 , wherein for the read command:
the network interface is responsive to the second commands to:
retrieve the at least two portions of data from the storage devices; and
store the retrieved at least two portions of data in local memory at the locations pointed to by the local payload pointers of the second commands; and
the Programmable Logic Device is configured to:
retrieve the at least two portions of data from the local memory at the locations pointed to by the local payload pointers of the first commands; and
restore the retrieved at least two portions of data responsive to the first commands to provide the data set.
8. A method for processing data, comprising:
providing a command and a payload pointer to an accelerator system;
obtaining, by the accelerator system, a data set responsive to the payload pointer;
generating at least two portions of data from the data set by the accelerator system responsive to the payload pointer;
storing the at least two portions of data locally in memory of the accelerator system;
initiating a memory access by a network interface of the accelerator system;
obtaining the at least two portions of data from the memory responsive to the memory access; and
transmitting the at least two portions of data obtained by the network interface to a cloud computing storage network, wherein the cloud computing storage network includes at least two respectively remote storage devices, and wherein the data set is restorable from at least a plurality of the at least two portions of data by recombining data from the plurality of the at least two portions of data.
9. The method of claim 8 , wherein:
the memory access is a direct memory access;
the direct memory access and the at least two portions of data obtained in response thereto are routed peer-to-peer by a switch coupling the network interface and Programmable Logic Device (“PLD”) to one another; and
the memory is coupled to the PLD.
10. The method of claim 9 , further comprising:
retrieving, using the network interface, the at least two portions of data from the cloud computing storage network; and
storing the retrieved at least two portions of data in the memory for subsequent processing by the PLD.
11. The method of claim 8 , wherein generating the at least two portions of data and transmitting the at least two portions of data to the cloud computing storage network is substantially transparent to the accelerator system.