Abstract: Exemplary methods, apparatuses, and systems determine that a cache is to be migrated from a first storage device to a second storage device. The cache includes cache entries organized in a first list of cache entries and a second list of cache entries. Only a portion of all cache entries from the first and second lists is selected for migration to the second storage device. The selected cache entries and metadata for cache entries from the first or second list that were not selected are migrated from the first storage device to the second storage device.

Abstract: Exemplary methods, apparatuses, and systems include a first host system configuring storage of the first host to serve as a primary cache for a virtual machine running on the first host. A second host system configures storage of the second host to serve as a secondary cache and boots a placeholder virtual machine. The first host transmits, in response to write operations from the virtual machine directed to the primary cache, copies of the write operations to the second host to create mirrored copies on the secondary cache. The first host acknowledges each write operation from the virtual machine when the write operation is committed to both the primary cache and the secondary cache. When the virtual machine is restarted on the second host in response to a failure or migration event, the secondary cache is promoted to serve as a new primary cache for the virtual machine.

Abstract: Techniques for enabling integration between a storage system and a host system that performs write-back caching are provided. In one embodiment, the host system can transmit to the storage system a command indicating that the host system intends to cache, in a write-back cache, writes directed to a range of logical block addresses (LBAs). The host system can further receive from the storage system a response indicating whether the command is accepted or rejected. If the command is accepted, the host system can initiate the caching of writes in the write-back cache.

Abstract: A software module capable of simultaneously supporting multiple services provides log message throttling for each service with a separate “per service” log message buffer. When the software module is a device driver, for example, each device controlled by the device driver is allocated a message buffer to store descriptive log messages. Upon generation of a warning log message, descriptive log messages in the message buffer are flushed to a log file for review by an administrator. Furthermore, log message throttling may be implemented by only flushing the message buffer upon certain occurrences of warning log messages, such as in accordance with an exponential back-off algorithm.

Abstract: Techniques for replicating a write-back cache are provided. In one embodiment, a first computer system can receive a write request from a virtual machine (VM) that includes data to be written to a shared storage device. The first computer system can further save the data in a local cache device and transmit, via a network interface controller (NIC), a data packet including the data to a second computer system. Upon determining that the data packet has been transmitted via the NIC, the first computer system can communicate an acknowledgment to the VM indicating that the write request has been successfully processed.

Abstract: Techniques are disclosed for reducing perceived read latency. Upon receiving a read request with a scatter-gather array from a guest operating system running on a virtual machine (VM), an early read return virtualization (ERRV) component of a virtual machine monitor fills the scatter-gather array with data from a cache and data retrieved via input-output requests (IOs) to media. The ERRV component is configured to return the read request before all IOs have completed based on a predefined policy. Prior to returning the read, the ERRV component may unmap unfilled pages of the scatter-gather array until data for the unmapped pages becomes available when IOs to the external media complete. Later accesses to unmapped pages will generate page faults, which are handled by stunning the VMs from which the access requests originated until, e.g., all elements of the SG array are filled and all pages of the SG array are mapped.

Abstract: Embodiments present a virtual disk image to applications such as virtual machines (VMs) executing on a computing device. The virtual disk image corresponds to one or more subparts of binary large objects (blobs) of data stored by a cloud service, and is implemented in a log structured format. Grains of the virtual disk image are cached by the computing device. The computing device caches only a subset of the grains and performs write operations without blocking the applications to reduce storage latency perceived by the applications. Some embodiments enable the applications that lack enterprise class storage to benefit from enterprise class cloud storage services.

Abstract: A cache is maintained with write order numbers that indicate orders of writes into the cache, so that periodic partial flushes of the cache can be executed while maintaining write order consistency. A method of storing data into the cache includes receiving a request to write data into the cache, identifying lines in the cache for storing the data, writing the data into the lines of the cache, storing a write order number, and associating the write order number with the lines of the cache. A method of flushing a cache having cache lines associated with write order numbers includes the steps of identifying lines in the cache that are associated with either a selected write order number or a write order number that is less than the selected write order number, and flushing data stored in the identified lines to a persistent storage.

Abstract: System and methods for safely sharing universal serial bus (USB) devices are provided. A request is received to switch from a state in which user mode drivers are not enabled to a state in which user mode drivers are enabled. For each kernel mode USB driver in the system, an I/O control (IOCTL) is called in each kernel mode USB driver. The IOCTL requests that the kernel mode USB driver yield devices on which it has a claim. Also, for each kernel mode USB driver in the system, it is determined if the kernel mode USB driver will yield one or more devices on which it has a claim. For each kernel mode USB driver that will yield and for each device claimed by the kernel mode USB driver for which the kernel mode USB driver respects a request to unclaim the device, a process is executed as if there had been a hot unplug event on the device.

Abstract: System and methods for safely sharing universal serial bus (USB) devices are provided. A request is received to switch from a state in which user mode drivers are not enabled to a state in which user mode drivers are enabled. For each kernel mode USB driver in the system, an I/O control (IOCTL) is called in each kernel mode USB driver. The IOCTL requests that the kernel mode USB driver yield devices on which it has a claim. Also, for each kernel mode USB driver in the system, it is determined if the kernel mode USB driver will yield one or more devices on which it has a claim. For each kernel mode USB driver that will yield and for each device claimed by the kernel mode USB driver for which the kernel mode USB driver respects a request to unclaim the device, a process is executed as if there had been a hot unplug event on the device.

Abstract: Described herein are systems and methods for preventing a user mode USB driver from performing IOCTL operations other than read-safe IOCTLs on a USB device that has been claimed by a kernel mode driver or is in use by another user mode USB driver. In one method, it is determined whether a kernel mode USB driver will claim a device or whether the device will be available to be claimed by user mode USB drivers. In the event the device is claimed by a kernel mode USB driver, user mode USB drivers will be prevented from claiming the device. In the event the device is available for use by user mode USB drivers, but has been opened for write by one user mode USB driver, all other user mode USB drivers will be prevented from claiming the device. All IOCTL operations other than read-safe IOCTLs will be prevented from being performed by a user mode USB driver unless that USB driver has claimed the device.

Abstract: One embodiment is a method of transferring data from a computer system to a Universal Serial Bus (USB) device after a computer system crash where interrupts are masked, the method comprising: (a) detecting the computer system crash; (b) transferring at least a portion of the data to a USB driver for the USB device; (c) the USB driver transferring the portion of the data to a USB controller driver for a USB controller for the USB device; (d) the USB controller driver causing the USB controller to transfer the portion of the data to the USB device; (e) polling the USB controller to determine whether the data transfer was completed; and (f) if the data transfer was completed, providing a notification to the computer system.

Abstract: Described herein are systems and methods for preventing a user mode USB driver from performing IOCTL operations other than read-safe IOCTLs on a USB device that has been claimed by a kernel mode driver or is in use by another user mode USB driver. In one method, it is determined whether a kernel mode USB driver will claim a device or whether the device will be available to be claimed by user mode USB drivers. In the event the device is claimed by a kernel mode USB driver, user mode USB drivers will be prevented from claiming the device. In the event the device is available for use by user mode USB drivers, but has been opened for write by one user mode USB driver, all other user mode USB drivers will be prevented from claiming the device. All IOCTL operations other than read-safe IOCTLs will be prevented from being performed by a user mode USB driver unless that USB driver has claimed the device.

Abstract: In one embodiment, a method includes transitioning control to a virtual machine (VM) upon receiving a request from a virtual machine monitor (VMM), determining that the request to transition control is associated with a request to be informed of an open event window, performing an event window check to determine whether an even window of the VM is open, and transitioning control to the VMM if the event window check indicates that the event window of the VM is open.

Abstract: A software module capable of simultaneously supporting multiple services provides log message throttling for each service with a separate “per service” log message buffer. When the software module is a device driver, for example, each device controlled by the device driver is allocated a message buffer to store descriptive log messages. Upon generation of a warning log message, descriptive log messages in the message buffer are flushed to a log file for review by an administrator. Furthermore, log message throttling may be implemented by only flushing the message buffer upon certain occurrences of warning log messages, such as in accordance with an exponential back-off algorithm.

Abstract: In one embodiment, a method includes transitioning control to a virtual machine (VM) from a virtual machine monitor (VMM), determining that a VMM timer indicator is set to an enabling value, and identifying a VMM timer value configured by the VMM. The method further includes periodically comparing a current value of a timing source with the VMM timer value, generating an internal event if the current value of the timing source has reached the VMM timer value, and transitioning control to the VMM in response to the internal event without incurring an event handling procedure in any one of the VMM and the VM.

Abstract: In one embodiment, a processor mode is provided for guest software. The processor mode enables the guest software to operate at a privilege level intended by the guest software. When the guest software attempts to perform an operation restricted by the processor mode, the processor mode is exited to transfer control over the operation to a virtual-machine monitor, which runs outside this processor mode.

Abstract: In some embodiments, the invention involves a system to deprivilege components of a virtual machine monitor and enable deprivileged service virtual machines (SVMs) to handle selected trapped events. An embodiment of the invention is a hybrid VMM operating on a platform with hardware virtualization support. The hybrid VMM utilizes features from both hypervisor-based and host-based VMM architectures. In at least one embodiment, the functionality of a traditional VMM is partitioned into a small platform-dependent part called a micro-hypervisor (MH) and one or more platform-independent parts called service virtual machines (SVMs). The micro-hypervisor operates at a higher virtual machine (VM) privilege level than any SVM, while the SVM and other VMs may still have access to any instruction set architecture (ISA) privilege level. Other embodiments are described and claimed.

Abstract: One embodiment is a method of transferring data from a computer system to a Universal Serial Bus (USB) device after a computer system crash where interrupts are masked, the method comprising: (a) detecting the computer system crash; (b) transferring at least a portion of the data to a USB driver for the USB device; (c) the USB driver transferring the portion of the data to a USB controller driver for a USB controller for the USB device; (d) the USB controller driver causing the USB controller to transfer the portion of the data to the USB device; (e) polling the USB controller to determine whether the data transfer was completed; and (f) if the data transfer was completed, providing a notification to the computer system.

Abstract: In one embodiment, a method includes transitioning control to a virtual machine (VM) upon receiving a request from a virtual machine monitor (VMM), determining that the request to transition control is associated with a request to be informed of an open event window, performing an event window check to determine whether an even window of the VM is open, and transitioning control to the VMM if the event window check indicates that the event window of the VM is open.