Abstract: Techniques for implementing dynamic base disk mirroring for linked clones are provided. In one set of embodiments, a first node in a distributed storage system can monitor a congestion level of a base disk residing on the first node, where the base disk is shared by a plurality of linked clones. Upon determining that the congestion level exceeds a threshold, the first node can send, to a second node, a request to create a mirror of the base disk on that second node. Upon receiving an acknowledgement from the second node that the mirror has been successfully created, the first node can update a mirror set associated with the base disk to include an entry identifying the mirror. The first node can then communicate the updated mirror set to one or more other nodes.

Abstract: Techniques for implementing dynamic base disk mirroring for linked clones are provided. In one set of embodiments, a first node in a distributed storage system can monitor a congestion level of a base disk residing on the first node, where the base disk is shared by a plurality of linked clones. Upon determining that the congestion level exceeds a threshold, the first node can send, to a second node, a request to create a mirror of the base disk on that second node. Upon receiving an acknowledgement from the second node that the mirror has been successfully created, the first node can update a mirror set associated with the base disk to include an entry identifying the mirror. The first node can then communicate the updated mirror set to one or more other nodes.

Abstract: Presented herein are methodologies for implementing erasure coding in a distributed log structured storage system. A method includes receiving a write request for first data from a file system, selecting a physical sector on a selected storage device in an array of storage devices on which to store the first data, assigning a key to the physical sector, storing the key and an indication of the physical sector in a key-to-physical medium map, erasure coding the data, including generating parity data associated with the first data, writing the first data and the parity data as a data stripe to each storage device in the array of storage devices and, in response to receiving the write request, sending the key to the file system. Read, update, and delete procedures in the context of a log structured framework are also described.

Abstract: Presented herein are methodologies for implementing erasure coding in a distributed log structured storage system. A method includes receiving a write request for first data from a file system, selecting a physical sector on a selected storage device in an array of storage devices on which to store the first data, assigning a key to the physical sector, storing the key and an indication of the physical sector in a key-to-physical medium map, erasure coding the data, including generating parity data associated with the first data, writing the first data and the parity data as a data stripe to each storage device in the array of storage devices and, in response to receiving the write request, sending the key to the file system. Read, update, and delete procedures in the context of a log structured framework are also described.

Abstract: Presented herein are methodologies for implementing erasure coding in a distributed log structured storage system. A method includes receiving a write request for first data from a file system, selecting a physical sector on a selected storage device in an array of storage devices on which to store the first data, assigning a key to the physical sector, storing the key and an indication of the physical sector in a key-to-physical medium map, erasure coding the data, including generating parity data associated with the first data, writing the first data and the parity data as a data stripe to each storage device in the array of storage devices and, in response to receiving the write request, sending the key to the file system. Read, update, and delete procedures in the context of a log structured framework are also described.

Abstract: Presented herein are methodologies for implementing erasure coding in a distributed log structured storage system. A method includes receiving a write request for first data from a file system, selecting a physical sector on a selected storage device in an array of storage devices on which to store the first data, assigning a key to the physical sector, storing the key and an indication of the physical sector in a key-to-physical medium map, erasure coding the data, including generating parity data associated with the first data, writing the first data and the parity data as a data stripe to each storage device in the array of storage devices and, in response to receiving the write request, sending the key to the file system. Read, update, and delete procedures in the context of a log structured framework are also described.

Abstract: A shared input/output (IO) resource is managed in a decentralized manner. Each of multiple hosts having IO access to the shared resource, computes an average latency value that is normalized with respect to average IO request sizes, and stores the computed normalized latency value for later use. The normalized latency values thus computed and stored may be used for a variety of different applications, including enforcing a quality of service (QoS) policy that is applied to the hosts, detecting a condition known as an anomaly where a host that is not bound by a QoS policy accesses the shared resource at a rate that impacts the level of service received by the plurality of hosts that are bound by the QoS policy, and migrating workloads between storage arrays to achieve load balancing across the storage arrays.

Abstract: A distributed storage system that performs automated load balancing is described. In an exemplary embodiment, a storage controller server determines if there is duplicative data in a distributed storage system. In this embodiment, the storage controller server detects a load balancing event in the distributed storage system, where the distributed storage system includes a plurality of virtual nodes distributed across a plurality of physical nodes. In response to detecting the load balancing event, the storage controller server determines that a current virtual node is to move from a source physical node to a destination physical node. In addition, the current virtual node is one of the plurality of virtual nodes and the source and destination physical nodes are in the plurality of physical nodes. The storage controller server further moves the current virtual node from the source physical node to the destination physical node.

Abstract: An anomaly in a shared input/output (IO) resource that is accessed by a plurality hosts or clients is detected when a host that is not bound by any QoS policy presents large workloads to a shared IO resource that is also accessed by hosts or clients that are governed by QoS policy. The anomaly detection triggers a response from the hosts or clients as a way to protect against the effect of the anomaly. The response is an increase in window sizes. The window sizes of the hosts or clients may be increased to the maximum window size or in proportion to their QoS shares.

Abstract: A shared input/output (IO) resource is managed in a decentralized manner. Each of multiple hosts having IO access to the shared resource, computes an average latency value that is normalized with respect to average IO request sizes, and stores the computed normalized latency value for later use. The normalized latency values thus computed and stored may be used for a variety of different applications, including enforcing a quality of service (QoS) policy that is applied to the hosts, detecting a condition known as an anomaly where a host that is not bound by a QoS policy accesses the shared resource at a rate that impacts the level of service received by the plurality of hosts that are bound by the QoS policy, and migrating workloads between storage arrays to achieve load balancing across the storage arrays.

Abstract: A shared input/output (IO) resource is managed in a decentralized manner. Each of multiple hosts having IO access to the shared resource, computes an average latency value that is normalized with respect to average IO request sizes, and stores the computed normalized latency value for later use. The normalized latency values thus computed and stored may be used for a variety of different applications, including enforcing a quality of service (QoS) policy that is applied to the hosts, detecting a condition known as an anomaly where a host that is not bound by a QoS policy accesses the shared resource at a rate that impacts the level of service received by the plurality of hosts that are bound by the QoS policy, and migration of workloads between storage arrays to achieve load balancing across the storage arrays.

Abstract: A system and method for maintaining processing order while permitting parallelism. Processing of a piece of work is divided into a plurality of stages. At each stage, a task advancing the work towards completion is performed. By performing processing as a sequence of tasks, processing can be done in parallel, with progress being made simultaneously on different pieces of work in different stages by a plurality of threads of execution.

Abstract: An anomaly in a shared input/ouput (IO) resource that is accessed by a plurality hosts or clients is detected when a host that is not bound by any QoS policy presents large workloads to a shared IO resource that is also accessed by hosts or clients that are governed by QoS policy. The anomaly detection triggers a response from the hosts or clients as a way to protect against the effect of the anomaly. The response is an increase in window sizes. The window sizes of the hosts or clients may be increased to the maximum window size or in proportion to their QoS shares.

Abstract: An anomaly in a shared input/ouput (IO) resource that is accessed by a plurality hosts or clients is detected when a host that is not bound by any QoS policy presents large workloads to a shared IO resource that is also accessed by hosts or clients that are governed by QoS policy. The anomaly detection triggers a response from the hosts or clients as a way to protect against the effect of the anomaly. The response is an increase in window sizes. The window sizes of the hosts or clients may be increased to the maximum window size or in proportion to their QoS shares.

Abstract: An embodiment of the present invention is directed to a method of determining a probable IP subnet of a network. The method includes maintaining a table of possible networks. The entries of the table include respective network addresses, respective subnet masks, and respective scores. The method also includes observing a packet, determining an IP address of the packet, and comparing the IP address of the packet with the respective network addresses of the entries of the table of possible networks. If the IP address of the packet falls under the respective network address of a particular entry, the respective score of the particular entry is adjusted. Otherwise, a new entry is created in the table of possible networks corresponding to the IP address of the packet. The method also includes ranking the entries of the table of possible networks based on their respective scores.

Abstract: In virtualized computer system in which a guest operating system runs on a virtual machine of a virtualized computer system and has direct access to a hardware device coupled to the virtualized computer system via a communication interface, a computer-implemented method of handling interrupts from the hardware device to the guest operating system includes: (a) receiving a physical interrupt from the hardware device on a shared interrupt line of an interrupt controller; (b) masking the shared interrupt line of the interrupt controller; (c) generating a virtual interrupt corresponding to the physical interrupt to the guest operating system; and (d) the guest operating system executing an interrupt service routine.

Abstract: In virtualized computer system in which a guest operating system runs on a virtual machine of a virtualized computer system and has direct access to a hardware device coupled to the virtualized computer system via a communication interface, a computer-implemented method of handling interrupts from the hardware device to the guest operating system includes: (a) receiving a physical interrupt from the hardware device on a shared interrupt line of an interrupt controller; (b) masking the shared interrupt line of the interrupt controller; (c) generating a virtual interrupt corresponding to the physical interrupt to the guest operating system; and (d) the guest operating system executing an interrupt service routine.

Abstract: A system and method for maintaining processing order while permitting parallelism. Processing of a piece of work is divided into a plurality of stages. At each stage, a task advancing the work towards completion is performed. By performing processing as a sequence of tasks, processing can be done in parallel, with progress being made simultaneously on different pieces of work in different stages by a plurality of threads of execution.