Vmware, Inc.

Emc Corporation

Zhaohui Guo

Jin Feng

Hong Li

Jian Zhao

Zhou Huang

Haitao Zhou

Wenguang Wang

Richard Paul Spillane

Junlong Gao

Philip C Love

Christos Karamanolis

Eric Shannon Knauft

Yuan Yuan Zhang

Ye Zhang

Ming Yang

Srinath Premachandran

Abhay Kumar Jain

Mounesh Badiger

Sriram Patil

Shugang Guo

Srinivas Aji

Mingzhou Joe Sun

Aditya Kotwal

Yixiang Zhang

Mario Wurzl

Yizheng Chen

Wangping He

A method is provided for a hyper-converged storage-compute system to implement an active-active failover architecture for providing Internet Small Computer System Interface (iSCSI) target service. The method intelligently selects multiple hosts to become storage nodes that process iSCSI input/output (I/O) for a target. The method further enables iSCSI persistent reservation (PR) to handle iSCSI I/Os from multiple initiators.

Active-active architecture for distributed ISCSI target in hyper-converged storage

One example method to perform object-based load balancing in a distributed storage system of a virtualization system supported by a cluster of host machines may include determining, by a first host machine in the cluster, whether any host machine in the cluster has affinity to a site. The method may also include determining, by the first host machine, whether to distribute affinity Internet small computer system interface (iSCSI) targets owned by the first host machine to at least the second machine based on a first balance objective associated with the site, and after having considered the first balance objective, determining, by the first host machine, whether to distribute iSCSI targets without affinity owned by the first host machine to other host machines in the cluster based on a second balance objective associated with the cluster.

Object-based load balancing approaches in distributed storage system

One example method to support Internet small computer system interface (iSCSI) services in a distributed storage system of a virtualization system may include receiving a notification of a change event associated with a global virtual Internet Protocol (VIP) object, wherein the global VIP object is associated with a virtual IP address for the iSCSI services. The method may also include in response to a first determination that the first host machine does not own the global VIP object and in response to a second determination that the first host machine gains ownership of the global VIP object based on the notification, locking the global VIP object and binding the virtual IP address to the first host machine.

Object-based approaches to support internet small computer system interface (ISCSI) services in distributed storage system

Certain Embodiments described herein relate to configuring the network-storage stack of two devices (e.g., physical or virtual) communicating together (e.g., an initiator and a target, as defined below) with Internet Small Computer Systems Interface (iSCSI) extension for remote direct memory access (RDMA) iSER, which is a protocol designed to utilize RDMA to accelerate iSCSI data transfer. The iSER protocol is implemented as an iSER datamover layer that acts as an interface between an iSCSI layer and an RDMA layer of the network-storage stacks of the two devices. Using iSER in conjunction with RDMA allows for bypassing the existing traditional network protocol layers (e.g., TCP/IP protocol layers) of the devices and permits data to be transferred directly, between the two devices, using certain memory buffers, thereby avoiding memory copies taking place when the existing network protocol layers are used.

Internet small computer interface systems extension for remote direct memory access (RDMA) for distributed hyper-converged storage systems

The disclosure herein describes enabling use of a logical unit for data storage in a distributed storage system using a plurality of backend data objects. Based on receiving instructions to create a logical unit of a logical unit size, a target backend object size to be used with the logical unit is determined, and a plurality of backend objects for allocation to the logical unit is calculated. The backend objects are allocated to the logical unit and a metadata mapping associated with the logical unit is generated. The metadata mapping associates logical block addresses of the logical unit to the allocated backend objects. The logical unit is linked with the metadata mapping in an input/output (I/O) service and, based on the linked metadata mapping, I/O traffic is routed to and from the logical unit. Using multiple backend objects enhances flexibility and efficiency of data storage on the distributed storage system.

Routing traffic of a logical unit to multiple backend data objects based on metadata mapping

System and method for writing updated versions of a configuration data file for a distributed file system in a storage system uses a directory renaming operation to write a new updated version of the configuration data file using the latest version of the configuration data file and a target directory. After the latest version of the configuration data file is modified by a particular host computer in the storage system, the modified configuration data file is written to a temporary file. The directory naming operation is then initiated on the temporary file to change the directory for the temporary file to the target directory. If the directory renaming operation has failed, a retry is performed by the particular host computer to write the new updated version of the configuration data file using a new latest version and a new target directory.

Lockless method for writing updated versions of a configuration data file for a distributed file system using directory renaming

Disclosed are a method and system for managing multi-threaded concurrent access to a cache data structure. The cache data structure includes a hash table and three queues. The hash table includes a list of elements for each hash bucket with each hash bucket containing a mutex object and elements in each of the queues containing lock objects. Multiple threads can each lock a different hash bucket to have access to the list, and multiple threads can each lock a different element in the queues. The locks permit highly concurrent access to the cache data structure without conflict. Also, atomic operations are used to obtain pointers to elements in the queues so that a thread can safely advance each pointer. Race conditions that are encountered with locking an element in the queues or entering an element into the hash table are detected, and the operation encountering the race condition is retried.

System and method of a highly concurrent cache replacement algorithm

Techniques are disclosed for using in-memory replicated object to support file services. Certain embodiments provide a method of storing persistent file handles in a storage system comprising a plurality of computing devices. The method may include requesting to write a persistent file handle corresponding to a file to a file system stored on the plurality of computing devices. The request may be translated to a block input/output (I/O) command to an in-memory object, the in-memory object representing at least a portion of the file system, a copy of the in-memory object being stored at each of the plurality of computing devices in volatile memory. The persistent file handle may then be written to the copy of the in-memory object stored in the volatile memory of each of the plurality of computing devices.

System and method of using in-memory replicated object to support file services wherein file server converts request to block I/O command of file handle, replicating said block I/O command across plural distributed storage module and performing said block I/O command by local storage module

A given host machine in a virtualization system having a virtual distributed storage system may receive an iSCSI protocol packet from a computer system separate from the given host machine. Processing the iSCSI protocol may include accessing distributed storage device (iSCSI target) comprising storage connected to the two or more host machines in the virtualization system. The given host machine may generate an outbound iSCSI protocol packet comprising return data received from the target and send the outbound iSCSI protocol packet to the computer system.

Distributed iSCSI target for distributed hyper-converged storage

A method to present information associated with a group of virtual components in a virtualized environment includes collecting a first resource usage data and a second resource usage data associated with a virtual component of the group of virtual components. In response to a region of the graphical representation associated with a difference between the first resource usage data and the second resource usage data exceeding a threshold, the method also includes collecting first graphical representation rendering information to render the first resource usage data as a first graphic in the graphical representation, collecting second graphical representation rendering information to render the second resource usage data as a second graphic in the graphical representation, and collecting third graphical representation rendering information including a reduction factor to reduce the region in the graphical representation to a third graphic in the graphical representation.

Graphical display for illustrating resource management

Virtualizing a service is disclosed. A request to access a service from a first server is received from a client. A secret data associated with the first server is used to process the received request. The processed request is sent to a second server. The first and second servers are associated with a virtualization; and wherein the processed request can be used by the second server to authenticate the client.

Authenticated service virtualization

The invention is a system and method for ensuring that a storage virtualization system performs as a trusted delegate for a user attempting to access data on an authenticated server. Data is moved from a first server to a second server by a storage virtualization system. Even though the user may have been authenticated to access the first server, the storage virtualization system which re-directs user requests to the migrated data on the second server may not be authenticated. The first time a user is re-directed to the second server, the storage virtualization system will obtain request to act as a trusted delegate for the user by communicating with a third server, such as a Kerberos domain controller. After the storage virtualization system is properly authenticated as a trusted delegate for the user, future re-directed requests will be seamless and transparent to the user.

Growing community of inventors

Shanghai, China

Zhaohui Guo