Emc Corporation

Emc IP Holding Company LLC

Orit Levin-Michael

Rong Yu

Yechiel Yochai

John W Lefferts

Scott Auchmoody

Scott H Ogata

Peng Yin

Qun Fan

Dan Aharoni

Stephen Richard Ives

Josef Ezra

Pei-Ching Hwang

Adnan Sahin

Amnon Naamad

Salil Dangi

Mark Chamness

Charles J Hickey

Lee W Sapiro

David Shadmon

Roderick Klinger

Stephen F Modica

William N Eagle

A method, article of manufacture, system and apparatus for reducing data transfer and backup duration using a client side dataset paging cache. Backups may be associated with given datasets. Loading the same dataset previous backups pages and champions into memory will improve backups performance.

Dataset paging cache for storage system

Protected data is divided into a plurality of segments. For each segment in the plurality of segments, a fingerprint is generated. The plurality of fingerprints is grouped into group(s) of fingerprints and for each group of fingerprints, a group fingerprint is generated. It is determined if the group fingerprint being processed is stored in a cache and if so, the group fingerprint being processed is flagged as one for which a virtual copy will be performed.

Client side cache for deduplication backup systems

A method, article of manufacture, apparatus, and system for a paging cache is disclosed. The backup cache may be broken into pages, and a subset of these pages may be memory resident. The pages may be sequentially loaded into memory to improve cache performance.

Paging cache for storage system

Exemplary methods for determining performance degradation of one or more disks at a storage system include analyzing disk performance statistics of a target storage system periodically collected during a storage of data from a source storage system to the target storage system, wherein the target storage system is a redundant array of inexpensive disks (RAID) storage system comprising of one or more arrays, each array comprising of a plurality of disks. The methods further include creating a predictive model based on the analysis of the collected performance statistics, and predicting, using the predictive model, that one or more disks of an array of the one or more arrays of the RAID storage system has degraded in performance.

Automated identification of the source of RAID performance degradation

Described are techniques for processing a data operation in a data storage system. A front-end component of the data storage system receives the data operation. In response to receiving the data operation, the front-end component performs first processing. The first processing includes determining whether the data operation is a read operation requesting to read a data portion which results in a cache miss; and if said determining determines that the data operation is a read operation resulting in a cache miss, performing read miss processing. Read miss processing includes sequential stream recognition processing performed by the front-end component to determine whether the data portion is included in a sequential stream.

Techniques for data prefetching

Described are techniques for processing a data operation in a data storage system. A front-end component receives the data operation to read a data portion. In response to receiving the data operation, the front-end component performs first processing. The first processing includes determining whether the data operation is a read operation resulting in a cache hit to a prefetched data portion of a sequential stream, and if said determining determines that said data operation results in a cache hit to a prefetched data portion, performing processing in connection with prefetching additional data for said sequential stream. The processing includes determining whether to prefetch additional data for said sequential stream and, if so, an amount of additional data to prefetch. The processing uses one or more criteria to determine one or more of an amount of data to prefetch in a single prefetch request and a track ahead parameter.

Techniques for determining an amount of data to prefetch

Data units are prefetched into a cache memory by executing a first prefetch task to prefetch a first prefetch series of data units from off-cache. A first prefetch operation is executed to prefetch and store a first selected set of data units. Decisions are made about the first prefetch task, subsequent prefetch operations and prefetch tasks based on in-cache data units associated with the first prefetch task and on other data units that are read from off-cache but that are not associated with the first prefetch task. A determination is made whether an additional data unit that is read from an off-cache logical location is a first prefetch series member. If so, it is associated with the first prefetch task, and, for decision making, is treated as having been prefetched into cache in accordance with the first prefetch task.

Methods and systems for incorporating sequential stream read requests into prefetch management for data storage having a cache memory

Adaptively pre-fetching data includes collecting a first set of statistics based on a number of avoidable read-misses in which data exists that is prior to data being read, collecting a second set of statistics based on a number of avoidable read-misses in which data exists that follows data being read, and collecting a third set of statistics based on said first and second sets of statistics. On the basis of the second set of statistics, a pre-fetch technique is selected from a first technique that pre-fetches data following data being read and a second technique that pre-fetches data before and following the data being read. The first and third set of statistics may be used to determine when to pre-fetch data.

Full track read for adaptive pre-fetching of data

Described are techniques for managing a sequential stream in a data storage system. A front-end component receives a plurality of data operations for a plurality of data portions and determines that the plurality of data portions are associated with a sequential stream. Each of the plurality of data portions of the sequential stream is associated with a sequential stream identifier, and a position indicator uniquely identifying a position of said each data portion in said sequential stream. Each of the plurality of data portions that is prefetched is associated with a remaining prefetched identifier and included in a prefetched chunk of data portions. The front-end component uses information about said sequential stream in connection with managing said sequential stream. The information includes the sequential stream identifier, one or more position indicators, and one or more remaining prefetched identifiers.

Techniques for management of information regarding a sequential stream

Described are techniques for determining a cache slot. A set of criteria for each of a plurality of families is received. A received data operation associated with a first of said plurality of families is obtained. It is determined, in accordance with the criteria associated with the received data operation, whether to allocate a cache slot in the cache for the received data operation. The criteria for the first family includes a minimum value and a maximum value used in determining a cache partition size range for the first family. The maximum value is used in determining a maximum cache partition size allowable for the first family.

Dynamic cache partitioning

A method for estimating the fall through time of a cache includes maintaining local statistics associated with each data cache slot replaced by a processor, reading from a memory global statistics related to each data cache slot replaced by a plurality of processors, calculating new global statistics, and estimating a fall through time of the cache in response to the global statistics at a predetermined rate. When the global statistics are the result of at least a threshold number of data cache slots having been replaced, the new global statistics are calculated by weighting the global statistics by a factor of less than one and adding the weighted global statistics to the local statistics. When the global statistics are not the result of at least a threshold number of data cache slots having been replaced, the new global statistics are calculated by adding a time related to the lapsed time to a global sum value associated with the time stamp difference of the replaced data cache slots. In a multi-processor computer system, each of the processors accesses the global memory to update the global statistics at a different time. The fall through time is estimated continuously at a first predetermined rate and the global statistics are updated at a second predetermined rate that, in one embodiment, is a multiple of the first predetermined rate.

Cache fall through time estimation

A system and method for determining whether activity graphs (hereinafter &ldquo;graphs&rdquo;) for logical volumes of data storage are correlated. A number of correlation criteria are applied to the graphs, including the area ratio criteria determines whether an area ratio described by the graph is at least as great as a predetermined value; the peak ratio criteria indicates that one graph is correlated to another graph if a peak ratio described by the two graphs is as large as a predetermined peak ratio value; and the sharp peak criteria indicates in the event that one graph is correlated to another graphs having at least one coincidental peak. Different combinations of all or some of the disclosed correlation criteria and the traditional correlation coefficient may be used to determine whether graphs of logical volumes are correlated, the level of correlation, and how &ldquo;strong&rdquo; of a correlation exists between the graphs.

Growing community of inventors

Irvine, CA, United States of America

Orit Levin-Michael