Emc Corporation

Emc IP Holding Company LLC

Nutanix, Inc.

Mark Chamness

Kai Chen

Eric Schnegelberger

Ping Chen

Michael John Dutch

Jun Tao

Christopher Hercules Claudatos

William Dale Andruss

Fabiano C Botelho

Michael Lee Federwisch

Guilherme Menezes

Rachel L Traylor

David Singer

Dmitry Serdyuk

Charles J Hickey

Orit Levin-Michael

Richard Mark Stern

Zhipeng Gong

Konstantin Tyapochkin

Revathi Anil Kumar

A first set of garbage collection (GC) features and non-GC features associated with a storage system are received, the first set of features being associated with a predetermined start date and a time window. A learning equation is generated having a plurality of vectors of GC features and a plurality of vectors of non-GC features. For a current iteration representing a current GC process, it is determined whether a first prior GC process was started within the time window. An entry of vectors of the non-GC features of the learning equation is populated based on corresponding feature values of the first set of non-GC features, in response to determining that the first prior GC process was started within the time window. A predetermined regression algorithm is applied to the learning equation to generate a GC duration predictive model to predict a GC duration of a subsequent GC process.

Method for prediction of the duration of garbage collection for backup storage systems

Examples of systems are described for calculating a probability of exceeding storage capacity of a virtualized system in a particular time period using probabilistic models. The probabilistic models may advantageously take variances of storage capacity into consideration.

Systems and methods for calculating a probability of exceeding storage capacity in a virtualized computing system

Retrieving information associated with a request for support is disclosed, including: receiving a request for customer support including geo-location data generated and included automatically in the request; and using the geo-location data to retrieve information associated with the request.

Geolocation for customer support mobile application

Diagnosis of defect(s) in a system is disclosed. A defect signature-based query is performed against system diagnostic data stored in one or more structured records. It is determined that a defect signature is associated with a system based at least in part on the query. Remediation information generated based at least in part on the defect signature and the system diagnostic data may be output.

Automated defect diagnosis from machine diagnostic data

A method for predicting disk failures in a RAID environment is provided. A data collection center receives hard disk status information from one or more sets of hard disks in a storage system. For each of the sets of hard disks, the data collection center calculates a transitional probability that a hard disk will fail within a predetermined period of time based on the hard disk status information, and generates a first risk profile for the hard disk based on the calculated transitional probability. The data collection center then generates a second risk profile for a set of hard disks based on two or more of the first risk profiles, and compares the second risk profiles of the sets of hard disks to determine which of the sets of hard disks has a highest probability of failing within the predetermined period of time.

Method for dynamically modeling medium error evolution to predict disk failure

A system for bug discovery using event reports comprises an interface and a processor. The interface is configured to receive symptom data extracted from event reports from a user system. The symptom data is stored in a symptom database. The symptom data comprises one or more symptoms each with a corresponding symptom occurrence time. The processor is configured to compose one or more SQL queries using one or more bug definitions and determine an existence of one or more bugs of the user system based at least in part on a result of querying the symptom database using the one or more SQL queries.

Set-based bugs discovery system via SQL query

A system for discovering bugs comprises an input interface and a processor. The input interface is configured to receive a bug definition. The bug definition includes a set operator with a time constraint and one or more set operands. The input interface is configured to receive an event log. The processor is configured to determine a list of bug occurrences using the bug definition and the event log.

System for discovering bugs using interval algebra query language

Performance information and configuration information is received for the plurality of computer systems. The computer systems are grouped into a plurality of clusters based at least in part on the performance information, where the plurality of clusters includes a first cluster and a second cluster. A system configuration associated with the first cluster is automatically identified from the configuration information and is automatically sent to the second cluster.

Automated defect and optimization discovery

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

Techniques for generating end-to-end sales leads based on storage capacity forecast using predictive modeling are described herein. According to one embodiment, diagnostic data is received from a data collector that periodically collects the diagnostic data from a storage system having one or more storage units to store data objects. A capacity forecaster coupled to the data collector forecasts a full capacity date using predictive modeling based on the diagnostic data, where the full capacity date estimates a date in which the one or more storage units reach a full storage capacity. A context generator coupled to the capacity forecaster generates a context having information identifying the one or more storage units of the storage system and an operator operating the storage system, wherein the context is used to communicate with the operator for acquiring an additional storage unit to increase storage capacity prior to the full capacity date.

Automated end-to-end sales process of storage appliances of storage systems using predictive modeling

Systems and methods are disclosed for predicting failure of a hard disk in a storage system. Embodiments are disclosed that predict failure of at least one hard disk in a storage system having a plurality hard disks. A data center reports to a data collection center than a hard disk has reported an initial non-zero medium error count (NMEC). The data collection center stores historic data of initial NMEC for many hard disks, and subsequent failure of those hard disks. From the historic data, the data collection center can report to the data center a prediction of when a hard disk reporting an initial NMEC may fail. Different models of hard disks fail at different times relative to a reported initial NMEC. The data collection center can track historic hard disk data by manufacturer, model of hard disk, and by model of storage system and thus can predict, by hard disk model, a probability of failure of a hard disk.

Prediction of disk failure

A system for managing a storage system comprises a processor and a memory. The processor is configured to receive storage system information from a deduplicating storage system. The processor is further configured to determine a capacity forecast based at least in part on the storage system information. The processor is further configured to provide a compression forecast. The memory is coupled to the processor and configured to provide the processor with instructions.

Capacity forecasting for a deduplicating storage system

Techniques for predicting data replication performance degradation are described. According to one embodiment, a method starts with receiving periodically data replication characteristics from a source storage and a target storage, the periodically received data replication characteristics being extracted from a replication process of the source storage replicating data to the target storage. The periodically received data replication characteristics are analyzed at a replication management server communicatively coupled with the source storage and the target storage, and a predictive modeling are applied. When the possibility of degradation of replication performance reaches a predetermined threshold, a notification is sent out.

Method and apparatus for predicting potential replication performance degradation

Defect remediation in a system is disclosed. A defect is mapped to one or more remediation actions, at least in part by using a processor to generate and perform a defect-based query against one or more structured remediation action records. A remediation package is generated using the processor. The remediation package may include the one or more remediation actions and data associated with the defect. The remediation package may be sent to a destination associated with the defect.

Defect remediation within a system

Mechanisms for predicting a GC duration are described herein. In one embodiment, the mechanisms include receiving a first set of features determined based on current operating status and prior garbage collection (GC) statistics of a first storage system. In one embodiment, the mechanisms include predicting a GC duration of a first GC process being performed at the first storage system by applying a predictive model on the first set of features, wherein the predictive model was generated based on a second set of features received periodically from a plurality of storage systems.

Growing community of inventors

Menlo Park, CA, United States of America

Mark Chamness