Sk Hynix Memory Solutions Inc.

Omnitier Storage, Inc.

Omnitier, Inc.

Derrick Preston Chu

Suneel Indupuru

Daryl Ng

Nandan Sridhar

Jason Bellorado

Marcus Marrow

Meng-Kun Lee

Nishant P Patil

Ameen Aslam

Priyanka Thakore

Prasanth Kumar

Prasanthi Relangi

An application and a plurality of types of storage in a distributed storage system are communicated with. A write instruction that includes a key-value pair that in turn includes a key and value is received from the application. The key-value pair is stored in a selected one of the plurality of types of storage where the selected type of storage is selected based at least in part on a size or access frequency of the key-value pair. A link to the stored key-value pair is stored, including by: generating a key hash based at least in part on the key from the key-value pair and selecting one of a plurality of rows in an extensible primary table in an index based at least in part on the key hash. If it is determined there is sufficient space, the link to the stored key-value pair is stored, in the selected row. If it is determined there is insufficient space, the key-value pair is stored in an overflow row in a secondary table.

Indexing technique for large scale distributed key-value systems

A plurality of pieces of write data are aggregated on a buffer to obtain a segment where the segment exceeds a smallest write size supported by storage. An address on the storage is determined for the segment. Location information and identifier(s) associated with the segment are recorded where the location information points to the storage, as opposed to the buffer, while the write data is being aggregated. When the write data has been aggregated into the segment, the segment is written to the storage wherein the location information remains unchanged in response to the writing to the storage.

Large segment management

Data to be stored is received from an application. A first type of storage, having a first set of storage properties, and a second type of storage, having a second set of storage properties, are selected from based on an access characteristic associated with the data, the first set of storage properties, and the second set of storage properties. The data is stored in the selected type of storage.

Application defined storage device

An indication is received that a word line has been read. The word line is part of a plurality of word lines (in solid state storage) which is divided into a plurality of groups. It is determined which group is associated with the read. A count of consecutive, at least potentially uninformative reads is updated based at least in part on the group associated with the read and a group associated with a prior read. It is determining if the count is greater than a threshold and in the event it is determined the count is greater than the threshold, a read disturb check is triggered.

Memory efficient triggers of read disturb checks in solid state storage

An access instruction which includes a logical block address (LBA) is received. A first-level table is accessed to obtain a first-level table entry associated with the LBA. From the first-level table entry, a location associated with a second-level table on solid state storage media is determined. The second-level table is accessed at the determined location to obtain a second-level table entry associated with the LBA. From the second-level table entry, a physical block address corresponding to the logical block address is determined.

Multi-level logical block address (LBA) mapping table for solid state

A plurality of write data is received and combined to obtain backup write data. The plurality of write data is stored in storage. On a storage controller, the plurality of write data is released without waiting for a confirmation associated with writing the plurality of write data to the storage. The backup write data is stored.

Recovering from a program failure by combining write data

A method of writing data to non-volatile computer storage is disclosed. A logical page of data is received and stored in an intermediate storage. A first portion of the logical page is read from the intermediate storage and written to a first physical page in the non-volatile computer storage. A second portion of the logical page is read from the intermediate storage and written to a second physical page in the non-volatile computer storage. A method of reading data from non-volatile computer storage is disclosed. A first portion of a logical page is read from a first physical page in the non-volatile computer storage and written in an intermediate storage. A second portion of the logical page is read from a second physical page and written in the intermediate storage. The first portion and the second portion of the logical page are concatenated to form the logical page.

Growing community of inventors

Santa Clara, CA, United States of America

Derrick Preston Chu