Hewlett Packard Enterprise Development LP

Hewlett-packard Development Company, L.p.

Narayanan Ananthakrishnan Nellayi

Kishore Kumar Muppirala

Sri Satya Sudhanva Kambhammettu

Subhakar Vipparti

Vijay Vishwanath Hegde

Anoop Kumar Raveendran

Bhanu Gollapudi Venkata Prakash

Shyamalendu Sarkar

Sumanesh Samanta

Satish Kumar Mopur

Senthil Kumar Ramakrishnan

Unnikrishnan Ponnan Katangot

Parthasarathi Ilangovan

Shankar Raman

Naveen B

Example implementation relates to a method for managing movement of set of entities associated with sampled data units in a cluster of nodes of a distributed storage system. A database is maintained for sampled data units received from the cluster of nodes. The method includes maintaining lineage information for the set of entities associated with the sampled data units received from each node. Based on the lineage information of a source node, a data set for migration is determined.

Selection and movement of data between nodes of a distributed storage system

Example implementation relates to generating a database for a set of entities associated with sampled data units of a storage system. A first data structure in the database maps, in a bit vector, each entity with a sampled signature of a set of sampled signatures. The set of sampled signatures are associated with the sampled data units. The set of entities associated with the set of sampled signatures are managed using the first data structure.

Representing and managing sampled data in storage systems

In one example, a method may include receiving a write operation corresponding to a portion of a data chunk stored at a first storage location in a write-in-place file system. The write-in-place file system may include encoded data chunks and unencoded data chunks. The method may include determining whether the data chunk is an encoded data chunk based on metadata associated with the data chunk, modifying the data chunk based on the write operation, and selectively performing a redirect-on-write operation on the modified data chunk based on the determination.

Selectively redirect-on-write data chunks in write-in-place file systems

An system for quota arbitration of a distributed file system (DFS) may obtain a current amount of storage used by a given quota entity on storage segments of the DFS, the given quota entity having a global quota for the DFS and each of the storage segments being assigned a respective first quota limit for the given quota entity. The system may determine a global available quota amount for the given quota entity across all the storage segments based on the obtained current amounts of storage used by the given quota entity and the global quota, and provision of a respective second quota limit for the given quota entity to each of the plurality of storage segments, based on a division of all or less than all of the global available quota amount.

Quota arbitration of a distributed file system

For storing data in a storage device, a storage allocation request may be received. The storage allocation request may include a logical offset of data to be stored. Further, a chunk size of the storage device and a device offset for a free region on the storage device may be received. An offset value may be computed based on the chunk size, file system block size, the device offset, and the logical offset. A device start address, for storing the data in response to the storage allocation request, can be determined by offsetting the device offset with the offset value.

Storing data in a storage device

A system and method for flexible device driver resource allocation is disclosed. In one embodiment, a method for allocating device driver resources in a data processing system includes statically allocating hardware resource pools to device drivers on a rigid basis during initialization of the data processing system, and dynamically altering the allocated hardware resource pools to the device drivers based on parameters associated with utilization of the allocated hardware resource pools during run-time.

System and method for flexible device driver resource allocation

A method and system for optimizing network I/O throughput is disclosed. In one embodiment, a method for optimizing an input/output (I/O) throughput for a storage network comprises measuring a service time for a storage device of the storage network in completing an I/O request serviced by a storage driver. The method also comprises determining a status of an I/O performance between the storage driver and the storage device by comparing the service time with an expected service time for the storage device in completing the I/O request, where the expected service time is calculated based on a type of the storage device and a size of the I/O request. The method further comprises adjusting a maximum queue depth associated with the storage device based on the status of the I/O performance.

Method and system for optimizing network input/output performance

The invention relates to a system and method for prioritizing one or more data processing operations in a computer storage system, the computer storage system including a plurality of modules, the method comprising receiving a command indicating one or more data processing operations to which priority is to be assigned and interfacing with each of the modules so as to prioritize the one or more data processing operations over other data processing operations.

Prioritising data processing operations

A method and system of a host device hosting multiple workloads for controlling flows of I/O requests directed to a storage device is disclosed. In one embodiment, a type of a response from the storage device reacting to an I/O request issued by an I/O stack layer of the host device is determined. Then, a workload associated with the I/O request is identified among the multiple workloads based on the response to the I/O request. Further, a maximum queue depth assigned to the workload is adjusted based on the type of the response, where the maximum queue depth is a maximum number of I/O requests from the workload which are concurrently issuable by the I/O stack layer.

Method and system of an I/O stack for controlling flows of workload specific I/O requests

A method and system for processing an input/output request on a multiprocessor computer system comprises pinning a process down to a processor issuing the input/output request. An identity of the processor is passed to a device driver which selects a device adapter request queue whose interrupt is bound to the identified processor and issues the request on that queue. The device accepts the request from the device adapter, processes the request and raises a completion interrupt to the identified processor. On completion of the input/output request the process is un-pinned from the processor. In an embodiment the device driver associates a vector of the identified processor with the request and the device, on completion of the request, interrupts the processor indicated by the vector.

Processing an input/output request on a multiprocessor system

The method, apparatus and system of an I/O forwarding technique for multi-interrupt capable I/O devices are disclosed. In one embodiment, a method of transferring an I/O request in a cache-coherent non-uniform memory access (ccNUMA) computer system including multiple cells (e.g., each cell may include multiple processors) that are connected via a system interconnect, includes receiving an I/O request from one of the multiple processors associated with one of the multiple cells in the ccNUMA computer system, associating a processor, corresponding to a multi-interrupt capable I/O interface that is servicing the I/O request, located in the one of the multiple cells as a lead processor, and executing an I/O initiation path and a completion path associated with the received I/O request on the lead processor upon associating the lead processor corresponding to the multi-interrupt capable I/O interface.

Growing community of inventors

Karnataka, India

Narayanan Ananthakrishnan Nellayi