Intel Corporation

Michael Mesnier

Arun Raghunath

Yi Zou

John Keys

Ian F Adams

Tsung-Yuan Charles Tai

Ren Wang

Feng Chen

Adam T Lake

Carl S Marshall

Anand S Ramalingam

Piotr Wysocki

Mariusz Barczak

Kapil Karkra

Daniel Robert McLeran

Weishuang Zhao

Francesc Guim Bernat

Kshitij Arun Doshi

Sanjeev N Trika

Ravishankar Iyer

Alexander W Min

John Joseph Browne

Susanne M Balle

Mesut A Ergin

Dave B Minturn

Scott D Hahn

Thijs Metsch

Luis Carlos Maria Remis

Paul Brett

Jonathan M Hughes

Nilesh N Shah

Bin Li

Adrian C Hoban

Wei Shen

Patrick Koeberl

Brennan A Watt

Tian Luo

Benjamin Boyer

Lay Wai Kong

Marcin Pioch

Veeraraghavan Ramamurthy

Mohammad Chowdhury

Richard Chuang

Various systems and methods for implementing computational storage are described herein. An orchestrator system is configured to: receive, at the orchestrator system, a registration package, the registration package including function code, a logical location of input data for the function code, and an event trigger for the function code, the event trigger set to trigger in response to when the input data is modified; interface with a storage service, the storage service to monitor the logical location of the input data and notify a location service when the input data is modified; interface with the location service to obtain a physical location of the input data, the location service to resolve the physical location from the logical location of the input data; and configure the function code to execute near the input data.

Computational storage in a function-as-a-service architecture

In one embodiment, a system comprises a host processor and a storage system. The storage system comprises one or more storage devices, and each storage device comprises a non-volatile memory and a compute offload controller. The non-volatile memory stores data, and the compute offload controller performs compute tasks on the data based on compute offload commands from the host processor.

Computational data storage systems

Techniques for storage and processing for distributed file systems are disclosed. In the illustrative embodiment, padding is placed between data elements in a file to be stored on a distributed file system. The file is to be split into several objects in order to be stored in the distributed file system, and the padding is used to prevent a data element from being split across two different objects. The objects are stored on data nodes, which analyze the objects to determine which data elements are present in the object as well at the location of those objects. The location of the objects is saved on the data storage device, and those locations can be used to perform queries on the data elements in the object on the data storage device itself. Such an approach can reduce transfer of data elements from data storage to local memory of the data node.

Technologies for storage and processing for distributed file systems

An embodiment of a semiconductor apparatus may include technology to selectively determine a set of data for background refresh based at least in part on host-provided information, and refresh the determined set of data on a persistent storage media as a background operation. Other embodiments are disclosed and claimed.

Selective background data refresh for SSDs

Technologies for managing data object requests in a storage node cluster include a proxy computing node communicatively coupled to the cluster of storage nodes. The proxy computing node is configured to receive data object requests from a communicatively coupled client computing device and identify a plurality of storage nodes of the cluster at which the data object of the data object request is stored. The proxy computing node is further configured to determine which of the identified storage nodes from which to retrieve the stored data object and transmit a request for the data object. Additionally, the proxy computing node is configured to estimate a request completion time based on a service time and a wait time for each of the identified storage nodes, as well as identify which of the storage nodes to retrieve the stored data object from based on the estimated request completion times. Other embodiments are described and claimed.

Technologies for managing data object requests in a storage node cluster

A system including a storage drive and a semiconductor apparatus coupled to the storage drive, is provided. The semiconductor apparatus may include one or more substrates and logic coupled to the one or more substrates, the logic coupled to the one or more substrates to: initiate managing resources of the storage drive and, if the storage drive loses capacity, determine an amount of capacity loss, create a reserved file that is associated with logical memory space in a file system, based on the amount of the capacity loss, and erase at least a portion of the reserved file so that logical memory space associated with an un-erased portion of the reserved file is usable by the storage drive.

Technology to manage capacity loss in storage drives

Embodiments of the present disclosure may relate to a compute offload controller that may include a parser to parse a received compute offload command, and identify a block-based compute descriptor based at least in part on the compute offload command. In some embodiments, the compute offload controller may further include an offload executor to perform an operation on data in a block-based storage device based at least in part on the block-based compute descriptor. In some embodiments, the block-based compute descriptor may include a virtual input object, a virtual output object, and a compute type identifier. Other embodiments may be described and/or claimed.

Compute offload in block storage using virtual objects

Technologies for managing replica caching in a distributed storage system include a storage manager device. The storage manager device is configured to receive a data write request to store replicas of data. Additionally, the storage manager device is configured to designate one of the replicas as a primary replica, select a first storage node to store the primary replica of the data in a cache storage and at least a second storage node to store a non-primary replica of the data in a non-cache storage. The storage manager device is further configured to include a hint in a first replication request to the first storage node that the data is to be stored in the cache storage of the first storage node as the primary replica. Further, the storage manager device is configured to transmit replication requests to the respective storage nodes. Other embodiments are described and claimed.

Managing replica caching in a distributed storage system

Technologies for reducing latency variation of stored data object requests include a proxy computing node communicatively coupled to a plurality of storage nodes. The proxy computing node is configured to determine whether to chunk a data object corresponding to a data object request received by the proxy computing node. Accordingly, the proxy computing node is configured to obtain a retrieval latency of the storage node and determine whether to chunk the data object based on the retrieval latency. The proxy computing node is further configured to, subsequent to a determination to chunk the data object, determine a chunk size (i.e., a portion of the data object) to be retrieved from the storage node and a remaining size of the data object at the storage node after the portion of the data object corresponding to the chunk request is received. Other embodiments are described and claimed.

Technologies for reducing latency variation of stored data object requests

Methods and apparatus related to framework and/or methodology for selective caching of Erasure Coded fragments in a distributed storage system are described. In one embodiment, a plurality of fragments of a data object are generated. Each of the plurality of fragments is Erasure Coded (EC) prior to storage at a storage node of a plurality of storage nodes. Each of the plurality of fragments is transmitted with a caching hint to indicate whether that fragment is to be cached at the storage node. Other embodiments are also disclosed and claimed.

Selective caching of erasure coded fragments in a distributed storage system

Examples may include techniques to monitor processing of I/O requests of an application being executed by a computing platform by collecting a trace of the I/O requests, the trace including an I/O class of each I/O request; replay the trace and automatically analyze possible cache configuration policies for using a cache during execution of the application by the computing platform; and determine an optimal cache configuration policy for the cache from the possible cache configuration policies. The optimal cache configuration policy may then be applied to use of the cache during subsequent execution of the application by the computing platform.

Cache utility modeling for automated cache configuration

Embodiments of apparatuses and methods for adaptive data compression and associated contextual information are described. In various embodiments, an apparatus may include a context monitoring module to gather contextual information for transmission of data and a policy module to gather user preference on cost associated with transmission of data. The apparatus may further include an analysis module to determine whether to compress data prior to transmission, based at least in part on the contextual information and the user preference. Other embodiments may be described and/or claimed.

Apparatus and methods for adaptive data compression

In accordance with some embodiments, classification of input/output requests from a database to a storage system may be performed. Each input/output request may be associated with a database class, and each database class may be mapped to a quality of service policy. Thus, quality of service may be enforced such that different data blocks within the storage system of the database may be afforded appropriate quality of service.

System and method for providing differentiated storage service in a database

Computer-readable storage media, computing devices and methods associated with file cache management are discussed herein. In embodiments, a computing device may include a file cache and a file cache manager coupled with the file cache. The file cache manager may be configured to implement a context-aware eviction policy to identify a candidate file for deletion from the file cache, from a plurality of individual files contained within the file cache, based at least in part on file-level context information associated with the individual files. In embodiments, the file-level context information may include an indication of access recency and access frequency associated with the individual files. In such embodiments, identifying the candidate file for deletion from the file cache may be based, at least in part, on both the access recency and the access frequency of the individual files. Other embodiments may be described and/or claimed.

Computing method and apparatus associated with context-aware management of a file cache

Growing community of inventors

Scappoose, OR, United States of America

Michael Mesnier