Microsoft Technology Licensing, LLC

David Matthew Kruse

Mathew George

James T Pinkerton

Thomas Ewan Jolly

Ahmed Hassan Mohamed

Roopesh C Battepati

Vladimir Petter

Neal Robert Christiansen

Sarosh Cyrus Havewala

Christian Gregory Allred

David Allen Dion

Lokesh Srinivas Koppolu

Balan Sethu Raman

Daniel Edward Lovinger

Prashanth Prahalad

Qibo Zhu

Ramanathan Ganapathy

Mingdong Shang

Mark Fishel Novak

Anders Bertil Vinberg

Niels Thomas Ferguson

Nir Ben-Zvi

Samartha Chandrashekar

Christopher Warner McCarron

Qiang Wang

Senthil Rajaram

Pradeep Madhavarapu

Yevgeniy Anatolievich Samsonov

Jeffrey Che Fuller

Amitabh Prakash Tamhane

John Anthony Messec

Nandagopal Kirubanandan

Sundar Subbarayan

Paul R Swan

Yun Lin

Henry Gabryjelski

Lars Reuther

Kevin Michael Broas

Thomas M Talpey

Venkataraman Ramanathan

Anthony Frederick Voellm

Kinshumann

Diaa E Fathalla

Gregory Robert Kramer

Tom Jolly

Iuliu Rus

Vladimirs Petters

Sriram Srinivasan

Shishir P Pardikar

Brian Scott Aust

Dustin G Friesenhahn

Michael C Johnson

Sanjay Anand

Thomas A Langan

Brian Kenneth Dewey

Kinshuman Kinshumann

Eileen C Brown

Matthew Stuart Garson

Thomas J Ootjers

Jay Paulus

Vikrant H Desai

MingDong (Gary) Shang

V Kishore Chintalapati

Sean Watson

Roopesh Battepati

Christopher Mccarron

Thomas E Jolly

Prahalad Prashanth

A generation identifier is employed with various systems and methods in order to identify situations where a workload has been reassigned to a new node and where a workload is still being processed by an old node during a failure between nodes. A master node may assign a workload to a worker node. The worker node sends a request to access target data. The request may be associated with a generation identifier and workload identifier that identifies the node and workload. At some point, a failure occurs between the master node and worker node. The master node reassigns the workload to another worker node. The new worker node accesses the target data with a different generation identifier, indicating to the storage system that the workload has been reassigned. The old worker node receives an indication from the storage system that the workload has been reassigned and stops processing the workload.

Distributed workload reassignment following communication failure

Examples described herein generally relate to a scalable multi-tier storage system. An entry may be added and/or deleted within the storage system. To delete an entry, the storage system may determine whether the entry corresponds to the file or the directory based on directory metadata, and request deletion of the directory metadata associated with the entry from the directory volume based on determining that the entry corresponds to the directory, and further requesting deletion of the file from a file volume based on a determination that the entry corresponds to the file. To add a file, the storage system may generate directory metadata associated with the entry in the directory volume based on a determination that the entry corresponds to the directory, and may further allocate file metadata in the file volume based on a determination that the entry corresponds to the file.

Scalable multi-tier storage structures and techniques for accessing entries therein

Examples described herein generally relate to maintaining a snapshot of a multitier distributed file system. A snapshot directory volume of a directory volume can be created at least in part by copying, to the snapshot directory volume, metadata of the directory volume. Based on a file access request, it can be determined whether the file has not been snapshotted. If not, a snapshot file record of the file, and associated with the snapshot directory volume.

Techniques for snapshotting scalable multitier storage structures

Systems and methods are disclosed for clients and servers operating in a scaled cluster environment. Efficiencies are introduced to the process of connecting a client to a clustered environment by providing the client with the ability to attempt a connection with multiple servers in parallel. Servers operating the in the clustered environment are also capable of providing persistent storage of file handles and other state information. Ownership of the state information and persistent handles may be transferred between servers, thereby providing clients with the opportunity to move from one server to another while maintaining access to resources in the clustered environment.

SMB2 scaleout

A virtual secure mode is enabled for a virtual machine operating in a computing environment that is associated with a plurality of different trust levels. First, a virtual secure mode image is loaded into one or more memory pages of a virtual memory space of the virtual machine. Then, the one or more memory pages of the virtual memory space are made inaccessible to one or more trust levels having a relatively lower trust level than a launching trust level that is used by a virtual secure mode loader to load the virtual secure mode image. A target virtual trust level is also enabled on a launching virtual processor for the virtual machine that is higher than the launching trust level.

Secure management of operations on protected virtual machines

Embodiments provide a method and system for transferring data between different computing devices. Specifically, a communication session is established between a first computing device and a second computing device. The communication session may be established using a first communication protocol. The first computing device creates a virtual memory object which is bound to one or more memory blocks of the first computing device. A path to the virtual memory object is generated and the path is transmitted to the second computing device using the communication session. The second computing device may then read or write data directly into/from the virtual memory object using a second communication protocol that is different from the first communication protocol. The data is written into and read from the virtual memory object using file system commands.

Efficient programmatic memory access over network file access protocols

A request identifier is employed with various systems and methods in order to enable an application to pause execution during a failure and subsequently resume execution and communication with a storage system once the failure is resolved. An application residing on a node may attempt to access target data on a storage system. The application sends a request to access the target data. The request may be associated with a request identifier that identifies the application. When the storage system experiences a failure, the application receives a failure notification. The application pauses execution as a result of the notification. When the node reestablishes a connection to the storage system, the application resumes execution using the same request identifier. If the resume request is granted, the application is guaranteed that the target data was not modified during the failure.

Safe data access following storage failure

The embodiments described herein generally relate to a protocol for implementing data operations, e.g., a version of SMB, atop RDMA transports. In embodiments, systems and methods use the protocol definition, which specifies new messages for negotiating an RDMA connection and for transferring SMB2 data using the negotiated communication. A new protocol message may include new header information to determine message size, number of messages, and other information for sending the SMB2 data over RDMA. The header information is used to accommodate differences in message size requirements between RDMA and SMB2. The SMB Direct protocol allows SMB2 data to be fragmented into multiple individual RDMA messages that a receiver may then logically concatenate into a single SMB2 request or SMB2 response. The SMB Direct protocol also may allow SMB2 to transfer application data via efficient RDMA direct placement and to signal the application data's availability when the transfer is complete.

Transporting operations of arbitrary size over remote direct memory access

Growing community of inventors

Kirkland, WA, United States of America

David Matthew Kruse