Pensando Systems Inc.

Allen Hubbe

Varagur Chandrasekaran

Balakrishnan Raman

Harinadh Nagulapalli

Murty Subba Rama Chandra Kotha

Shrikant Jay Vaidya

Krishna Doddapaneni

Chaitanya Huilgol

J Bradley Smith

Varada Raja Kumar Kari

Andrew Boyer

Nitish Bhat

SR-IOV (single root IO virtualization) capable PCIe devices can implement virtual functions (VFs) that are assigned to VMs running on a host machine, thereby speeding IO operation by writing directly to the VMs' memory while bypassing the hypervisor managing the VMs. As such, VFs thwart the dirty page tracking that hypervisors use to minimize VM downtime when the VM is migrated between hosts. The SR-IOV PCIe devices can help resolve this problem by maintaining dirty page tracking data for VMs running on the host machine. The SR-IOV PCIe devices bypassing the hypervisor while writing into a memory page of the VM can set the dirty page tracking data to indicate the memory pages that are dirty (i.e., written to by the VF), and can provide access to the dirty page tracking data. The hypervisor can thereby obtain and use the dirty page tracking data.

Methods and systems for using a peripheral device to assist virtual machine IO memory access tracking

Tenants in data centers may want access to high precision clocks without having to run their own PTP stacks or reference clocks. Furthermore, different tenants may want their workloads synchronized to their own secured clock domain. PTP, the currently dominant synchronization protocol, allows for only 256 clock domains (CDs). Virtual CDs (vCDs) virtualize the concept of clock domains by maintaining a hardware clock within a host computer, receiving a network clock domain packet that includes a clock domain identifier and an origin timestamp produced by a reference clock, using the network clock domain packet to synchronize the hardware clock to the reference clock, and using the hardware clock to provide a hardware timestamp value to a virtual machine (VM) running on the host computer or to a process running on the host computer, wherein the hardware clock is secured from manipulation by the VM or by the process.

Methods and systems for providing a distributed clock as a service

Data centers often run long lived services such as web servers that are intended to run for hours, days, or even longer before being torn down and replaced with another instance of the long-lived service. Currently, many applications are being implemented with microservice architectures that run short lived services that start up, implement an operation, and are then torn down. An aspect of starting up a service is creating administrative data structures such as InfiniBand queue pairs. A packet processing pipeline having a DMA output stage can be configured to create the administrative data structures, thereby increasing the rate at which the administrative data structures are created. As a result, services running in data centers can be started up more rapidly and efficiently.

Methods and systems for using a packet processing pipeline to accelerate InfiniBand administrative operations

Inbound packets can be received by a network device that determines a receive pipeline latency metric based on a plurality of receive pipeline residency times of the inbound packets and determines a receive queue latency metric based on a plurality of receive queue residency times of the inbound packets. The receive queue latency metric and the receive pipeline latency metric can be reported to a data collector. The network appliance may also receive a plurality of outbound packets on a transmit queue, determine a transmit queue latency metric based on the transmit queue residency times of the outbound packets, and determine a transmit pipeline latency metric based on the transmit pipeline residency times of the outbound packets. The outbound packets may be transmitted toward their destination. The transmit queue latency metric and the transmit pipeline latency metric can be reported to the data collector.

Methods and systems for queue and pipeline latency metrology in network devices and smart NICs

Growing community of inventors

Durham, NC, United States of America

Allen Hubbe