International Business Machines Corporation

Intel Corporation

Cisco Technology, Inc.

Jiang Yu

Mengze Liao

Ziye Yang

Yadong Li

Anjali Singhai Jain

Daniel Daly

Ping Yu

Cunming Liang

Changpeng Liu

Liang J Ma

Jingjing Wu

Shaopeng He

Bo Cui

Hongjun Ni

Min Fang

Zhiguo Wen

Yang Liu

Jian Tang

Yufei Li

Yuefeng Jiang

Lijian Yang

Qing Fa Wang

Yu Sun

Yong Qing Lu

Da Li Liu

Wei Yin

Duo Chen

Shu Chen

Peng Lm Shao

Si Cong Zhou

Haitao Kang

Jinyi Pu

Chun Yao

Hong Xin Hou

Shu Luo

Wen Jun Jin

Qing Chuan Sun

Techniques for improving options templates are described. The techniques may be performed at least in part by a network edge device and may include receiving, from a network collector device, a request for an options template download, the request indicating whether a full options template or partial options template are requested. Based at least in part on whether a full options template or a partial options template are requested, the method may also include, transmitting a response to the network collector device including a segment of the options template, wherein each segment includes a flag indicating whether the segment is a last segment of at least one more segment will follow. Finally, the method may include receiving from the collector device and in response to transmitting a segment having the flag indicating the segment is a last segment, an acknowledgement that the full options template has been received.

Options template transport for software defined wide area networks

Examples described herein relate to a switch comprising a programmable data plane pipeline, wherein the programmable data plane pipeline is configured to provide microservice-to-microservice communications within a service mesh. In some examples, to provide microservice-to-microservice communications within a service mesh, the programmable data plane pipeline is to perform a forwarding operation for a communication from a first microservice to a second microservice. In some examples, to perform a forwarding operation for a communication from a first microservice to a second microservice, the programmable data plane pipeline is to utilize a reliable transport protocol.

Service mesh offload to network devices

Examples described herein relate to offload reliable transport management to a network interface device and store packets to be resent, based on received packet receipt acknowledgements (ACKs), into one or more kernel space queues that are also accessible in user space.

Reliable transport offloaded to network devices

Techniques for improving options templates for network traffic monitoring and analysis, using pull mode by a network collector device, and sending an acknowledgement, by the network collector device that the download was successfully received are described. The techniques may include transmitting, by a network collector device and to a network edge device, a request to download a full options template, receiving, by the network collector device, responses from the network edge device, each response including a segment of the full options template and each segment including a last segment flag indicating whether the segment is a last segment, and in response to receiving a segment having the last segment flag set, transmitting, by the network collector device and to the network edge device, an acknowledgement that the full options template has been received.

Examples described herein relate to a reliable transport protocol for packet transmission using an Address Family of an eXpress Data Path (AF_XDP) queue framework, wherein the AF_XDP queue framework is to provide a queue for received packet receipt acknowledgements (ACKs). In some examples, an AF_XDP socket is to connect a service with a driver for the network device, one or more queues are associated with the AF_XDP socket, and at least one of the one or more queues comprises a waiting queue for received packet receipt ACKs. In some examples, at least one of the one or more queues is to identify one or more packets for which ACKs have been received. In some examples, the network device is to re-transmit a packet identified by a descriptor in the waiting queue based on non-receipt of an ACK associated with the packet from a receiver.

Embodiments include a load calculator, a workload analyzer, and a decision module. The load calculator generates an input/output (IO) record for an asset. The IO record includes a count of read operations and write operations corresponding to the asset from each of a plurality of sites. The workload analyzer collects the IO record and generates at least one of a write threshold and a read threshold. The decision module generates a request for at least one of a promotion and a copy of the asset in response to a determination that an operation has reached at least one of the write threshold and the read threshold for the asset.

Network asset management

Embodiments for server port virtualization for guest logical unit number (LUN) masking in a host direct attach configuration using a storage adapter in a computing environment by a processor. An F switch port is simulated by an N storage port to enable either N-port virtualization (NPV) or N-port identification (ID) virtualization (NPIV) in the host direct attach configuration by directly attaching the N server port to the N storage port. A domain name system (DNS) operation is performed to cause each virtualized N-port ID to be mapped to fiber channel (FC) IDs in domain format of domain, area, port.

Server port virtualization for guest logical unit number (LUN) masking in a host direct attach configuration

Embodiments relate to a system, program product, and method for expanding the retention capacity of a write cache on a host bus adaptor (HBA) device through the use of non-volatile dual in-line memory modules (NVDIMMs). At least a portion of the existing NVDIMM capacity is leveraged to temporarily retain data that is to be written to persistent storage to reduce the potential for data loss during external electric power disruptions. A NVDIMM HBA Write Cache Module (NHWCM) facilitates the data file traffic transmission and I/O command generation and transmission between a processing device, the HBA device, the NVDIMMs, and the persistent storage.

Expansion of HBA write cache using NVDIMM

Aspects of the invention include receiving a request to transfer data from a first storage device, coupled to a sending server, to a second storage device, coupled to a receiving server. The data is transferred from the first storage device to the second storage device in response to the request. The transferring includes allocating a first temporary memory on the sending server and moving the data from the first storage device to the first temporary memory. The transferring also includes initiating a remote direct memory access (RDMA) between the first temporary memory and a second temporary memory on the second server. The RDMA causes the data to be transferred from the first temporary memory to the second temporary memory independently of an operating system executing on a processor of the sending server or the receiving server. The transferring further includes receiving a notification that the transfer completed.

Efficient silent data transmission between computer servers

A computer receives, at a target node, a replication target update, wherein the replication target update comprises a write frequency of a replication target and a size of the replication target. The computer creates a replication task to perform replication based upon the replication target update. The computer calculates a priority value of the replication task wherein the priority value is based upon the write frequency of the replication target and the size of the replication target. The computer inserts the replication task into a priority queue. The computer processes the replication task.

Data replication priority management

An approach for efficient booting. The approach loads a firmware into a memory from one or more integrated memories, wherein the one or more integrated memories are flash memories on a motherboard. The approach loads an operating system into the memory from the one or more integrated memories. Responsive to loading the operating system into the memory, the approach mounts one or more root filesystems.

Efficient booting system

Disclosed aspects relate to facilitating system design based on unified chip specification. It can be determined based on the system design that a first interface of a first chip is to be connected to a second interface of a second chip. Then a first configuration of the first interface and a second configuration of the second interface are determined based on a unified specification. The unified specification at least specifies configurations of a plurality of chip interfaces for respective usages. A hardware design may be automatically generated based on the first and second configurations. The hardware design may include a hardware-level connection between the first and second interfaces.

System design based on unified chip specification

A page replacement algorithm is provided. An idle range of memory pages is determined based, at least in part, on indications of references to memory pages in the idle range of memory pages, wherein the idle range of memory pages is a set of one or more memory pages. A first memory page is identified in the idle range of memory page for paging out of memory. The first memory page is identified based, at least in part, on indications of modifications to the memory pages. The first memory page is paged out of memory.

Growing community of inventors

Shanghai, China

Jiang Yu