International Business Machines Corporation

Vmware, Inc.

Keerthi B Kumar

Rajesh Venkatasubramanian

Pratap Subrahmanyam

Marcos Kawazoe Aguilera

Pramod Kumar

Swetha N Rao

Shailaja Mallya

Sairam Veeraswamy

Chetan L Gaonkar

Halesh Sadashiv

Kiran Tati

Nikhil Hegde

Juan M Casas, Jr

Uma Maheswara R Chandolu

Nitin Madhusudan Agrawal

Pramod Kumar P

Shubham Shashikant Patil

Kiran Dikshit

Suma M B Bhat

Brahadambal Srinivasan

Vishal R Mansur

Irfan Ulla Khan

Saurabh K Gupta

Jayalekshmy Nair

Arun Ramanathan

Supriya Jagadeesh

Puspanjali Kar

Manish Mishra

Deepak L Ranganath

Frederic Durairaj Maria Joseph

Disclosed are various embodiments for improving resiliency and performance of clustered memory. A computing device can acquire a chunk of byte-addressable memory from a cluster memory host. The computing device can then identify an active set of allocated memory pages and an inactive set of allocated memory pages for a process executing on the computing device. Next, the computing device can store the active set of allocated memory pages for the process in the memory of the computing device. Finally, the computing device can store the inactive set of allocated memory pages for the process in the chunk of byte-addressable memory of the cluster memory host.

Resiliency and performance for cluster memory

Disclosed are various embodiments for improving the resiliency and performance for clustered memory. A computing device can mark a page of the memory as being reclaimed. The computing device can then set the page of the memory as read-only. Next, the computing device can submit a write request for the contents of the page to individual ones of a plurality of memory hosts. Subsequently, the computing device can receive individual confirmations of a successful write of the page from the individual ones of the plurality of memory hosts. Then, the computing device can mark the page as free in response to receipt of the individual confirmations of the successful write from the individual ones of the plurality of memory hosts.

Disclosed are various embodiments for improving the resiliency and performance of clustered memory. A computing device can generate at least one parity page from at least a first local page and a second local page. The computing device can then submit a first write request for the first local page to a first one of a plurality of memory hosts. The computing device can also submit a second write request for the second local page to a second one of the plurality of memory hosts. Additionally, the computing device can submit a third write request for the parity page to a third one of the plurality of memory hosts.

Techniques for implementing RDMA-based recovery of dirty data in remote memory are provided. In one set of embodiments, upon occurrence of a failure at a first (i.e., source) host system, a second (i.e., failover) host system can allocate a new memory region corresponding to a memory region of the source host system and retrieve a baseline copy of the memory region from a storage backend shared by the source and failover host systems. The failover host system can further populate the new memory region with the baseline copy and retrieve one or more dirty page lists for the memory region from the source host system via RDMA, where the one or more dirty page lists identify memory pages in the memory region that include data updates not present in the baseline copy. For each memory page identified in the one or more dirty page lists, the failover host system can then copy the content of that memory page from the memory region of the source host system to the new memory region via RDMA.

Remote direct memory access (RDMA)-based recovery of dirty data in remote memory

Disclosed are various embodiments for improving the resiliency and performance of cluster memory. First, a computing device can submit a write request to a byte-addressable chunk of memory stored by a memory host, wherein the byte-addressable chunk of memory is read-only. Then, the computing device can determine that a page-fault occurred in response to the write request. Next, the computing device can copy a page associated with the write request from the byte-addressable chunk of memory to the memory of the computing device. Subsequently, the computing device can free the page from the memory host. Then, the computing device can update a page table entry for the page to refer to a location of the page in the memory of the computing device.

Copy and restore of page in byte-addressable chunks of cluster memory

Techniques for determining reliability of a workload migration activity are disclosed. In one embodiment, sub-tasks associated with the workload migration activity may be determined. Further, statistical data associated with an execution of the sub-tasks corresponding to different instances of the workload migration activity may be retrieved. Furthermore, a reliability model may be trained through machine learning using the statistical data to determine reliability of the workload migration activity. Then, the reliability of a new workload migration activity may be determined using the trained reliability model.

Reliability determination of workload migration activities

The disclosure provides an approach for creating a pool of memory out of local memories of host machines, and providing that pool for the hosts to use. The pool is managed by a controller that keeps track of memory usage and allocated memory among hosts. The controller allocates or reclaims memory between hosts, as needed by the hosts. Memory allocated from a second host to a first host may then be divided into smaller portions by the first host, and further allocated to virtual machines executing within the first host.

Remote memory in hypervisor

A method for transferring memory pages to a first and a second page repository identifies pages in a memory sharing operation for transfer to a first page repository and pages in a memory migration operation for transfer to a second page repository. Pages in the memory migration operation may be prepared for transfer prior to transfer of the pages in the memory sharing operation. Transferring pages in the migration operation may remove the need to transfer pages in the memory sharing operation.

Prioritizing pages to transfer for memory sharing

A method for improving accuracy of high resolution timers in a live partition migration comprises determining a status of each page of a plurality of pages to be copied from a source partition to a target partition during the live partition migration. One or more pages of the plurality of pages correspond to a respective high resolution timer. The method comprises determining that each of the one or more pages corresponding to a respective high resolution timer has a clean status. A clean status indicates that the corresponding page has been copied to the target partition and has not been modified subsequent to being copied to the target partition. The method also comprises halting operation of the source partition and initiating operation of the target partition in response to determining that each of the one or more pages corresponding to a respective high resolution timer has a clean status.

High resolution timer expiry in live partition migration

An example method of provisioning a virtual machine in a cloud system includes: receiving a request to provision the virtual machine in the cloud system; provisioning a first virtual machine managed by a hypervisor executing on a host computer in the cloud system, the hypervisor allocating resources to the first virtual machine; provisioning a nested hypervisor in the first virtual machine; and provisioning the virtual machine to execute on the nested hypervisor without re-virtualizing the resources allocated by the hypervisor to the nested hypervisor.

Nested hypervisor to run virtual machines on public clouds

Growing community of inventors

Bengaluru, India

Keerthi B Kumar