Dell Products Inc.

Emc IP Holding Company LLC

Jenny Derzhavetz

Vladimir Shveidel

Ronen Gazit

Dror Zalstein

Oran Asher Baruch

Bar David

Nimrod Shani

Oran Baruch

Michael Litvak

Yuri Chernyavsky

Techniques for using data logs with bonded page descriptors (PDs) in storage systems. The techniques include, in response to a write request from a storage client, performing, in a log structured data log, a single object allocation for a bonded page descriptor (PD)-page buffer (PB) pair, storing and persisting, in a PB of the bonded PD-PB pair, user data specified by the write request, and storing and persisting, in a PD object of the bonded PD-PB pair, metadata related to the user data. The techniques include, once the user data and related MD are stored and persisted in the bonded PD-PB pair, sending an acknowledgment message to the storage client, and de-staging, flushing, or transferring, in the background, the user data and/or related MD from the bonded PD-PB pair to a storage object. By performing a single object allocation, processing costs related to performing the object allocation can be reduced.

Data log with bonded descriptors

A searchable metadata change log is maintained in a persistent memory of a data storage system, and stores metadata changes to be aggregated and applied to data storage system metadata located in non-volatile data storage. New metadata changes are added to the metadata change log by storing the new metadata changes into the persistent memory, storing structural modifications to the metadata change log needed to add the new metadata changes to the metadata change log into a transaction entry located in the persistent memory, and processing the resulting transaction entry to perform the structural modifications to the metadata change log that were stored in the transaction entry.

Managing changes to metadata in a data storage system

A method, computer program product, and computing system for sensing a failure within a system within a computing device. The system may include a cache memory system and a vaulted memory comprising a random access memory (RAM) having a plurality of independent persistent areas. A primary node and secondary node may be provided. The primary node may occupy a first independent persistent area of the RAM of the vaulted memory. The secondary node may occupy a second independent persistent area of the RAM of the vaulted memory. Data within the vaulted memory may be written to a persistent media using an iterator. The data may include at least one dirty page. Writing data within the vaulted memory to the persistent media may include flushing the at least one dirty page to the persistent media.

Storage system and method using persistent memory

A technique for managing reference-count decrements (decrefs) includes defining an aggregation domain that includes multiple tablets created at respective points in time, each of the tablets containing multiple sorted buckets arranged in a respective sequence of bucket identifiers that is repeated across all of the tablets, the sorted buckets containing decrefs for respective sets of metadata pages. The technique further includes assigning a decref job to a specified set of bucket identifiers and destaging, by the decref job, multiple decrefs from the tablets to persistent metadata pages, beginning with decrefs in buckets having the specified set of bucket identifiers and proceeding until a target number of decrefs has been destaged.

Managing decref processing from tablets containing sorted buckets

In at least one embodiment, processing can include: receiving metadata updates for MD pages; recording the MD updates for the MD pages in a MD log; and performing a first destaging phase that writes the MD updates from the MD log to a tablet stored on non-volatile storage. The tablet can include a bucket area and an extension area. The bucket area can include buckets each of which is a predetermined bucket size. The first destaging phase can include: writing a portion of the MD updates from the MD log to the bucket area; and writing a second portion of the MD updates from the MD log to the extension area. Each MD update of the second portion can overflow a corresponding one of the buckets which is included in the bucket area and which is associated with one of the MD pages updated by the MD update.

Techniques using a variable size extension area with two phase destaging

A technique for managing metadata changes in a storage system provides different aggregation policies for use with different metadata. The technique includes assigning metadata changes to respective aggregation policies and storing the assigned metadata changes in a set of data structures. The technique further includes destaging the metadata changes from the set of data structures separately for the different aggregation policies in accordance with settings specific to those aggregation policies.

Processing metadata changes using multiple aggregation policies

In at least one embodiment, processing can include: determining that an inactive decref (decrement reference count) MDL (metadata log) of decref MD (metadata) updates stored in memory is unable to be destaged to a decref tier of non-volatile storage in a first destage phase of a two phase destage process; and responsive to determining that the inactive decref MDL is unable to be destaged to the decref tier, performing bypass destage processing of the inactive decref MDL, wherein said bypass destage processing includes directly applying MD updates of the inactive decref MDL to corresponding MD pages stored persistently in a MD page store and wherein said bypass destage processing omits storing the inactive decref MDL on the decref tier.

Bypass destaging of decrement reference count operations with delta log based architecture

A method, computer program product, and computing system for receiving a flush request for a metadata page stored in a storage array of a multi-node storage system. The flush request may be queued on a flush request lock queue on at least one node of the multi-node storage system. One or more flush requests may be processed, via multiple nodes of the multi-node storage system, on the metadata page based upon, at least in part, the flush request lock queue.

System and method for lockless destaging of metadata pages

A method, computer program product, and computing system for assigning a plurality of unique sequential identifiers to a plurality of tablets in a cache memory system. One or more metadata deltas associated with a metadata page stored in a storage array may be written to the plurality of tablets in the cache memory system. Each metadata delta stored in at least one tablet of the plurality of tablets may be written to the metadata page stored in the storage array, thus defining one or more destage tablets. A largest unique sequential identifier from the plurality of unique sequential identifiers assigned to the one or more destage tablets, may be written to the storage array, thus defining a current tablet identifier for the metadata page.

System and method for lockless reading of metadata pages

A method, computer program product, and computing system for receiving, at a local node, a request to buffer data on a remote persistent cache memory system of a remote node. A target memory address within the remote persistent cache memory system may be sent from the local node via a remote procedure call (RPC). The data may be sent from the local node to the target memory address within the remote persistent cache memory system via a remote direct memory access (RDMA) command.

System and method for multi-node buffer transfer

Techniques for providing hiccup-less failback and journal recovery. The techniques include determining a head position of a data log of a primary node and placing a marker at a corresponding position in a data log of a secondary node. In response to an IO request, the head of the primary log is moved to allocate space for writing a data entry, and the secondary node is directed to move a head of the secondary log to allocate space for writing the data entry. In response to a data entry being flushed from the primary node, a tail of the primary log is moved to reclaim allocated space for a previous data entry, and the secondary node is directed to move a tail of the secondary log to reclaim previously allocated space. In response to a tail of the secondary log coinciding with the marker position, content of the logs is synchronized.

Growing community of inventors

Raanana, Israel

Jenny Derzhavetz