Oracle International Corporation

Santosh Narayan Shilimkar

Jagwinder Singh Brar

David Dale Becker

Bryce Eugene Bockman

Ravi Shankar Thammaiah

Nagarajan Muthukrishnan

Sumanta Kumar Chatterjee

Shane Baker

Binoy Sukumaran

Steven Chervets

Lukasz Sulek

Jacob Robert Uecker

Raman Kumar Sonkhla

Vijayendra Lakkundi

Shankar Venugopal

Marcin Jakub Zablocki

Techniques and apparatus for data networking are described. In one example, a method of queuing Remote Direct Memory Access (RDMA) packets includes receiving a first RDMA packet having a first quality-of-service (QoS) data field; based on a value of the first QoS data field, queueing the first RDMA packet in a first queue of a plurality of queues; receiving a second RDMA packet having a second QoS data field; and based on a value of the second QoS data field, queueing the second RDMA packet in a second queue of the plurality of the queues, the second queue being different than the first queue.

Class-based queueing for scalable multi-tenant RDMA traffic

Techniques and apparatus for data networking are described. In one example, a method includes receiving a first Layer-2 Remote Direct Memory Access (RDMA) packet which includes a virtual local area network (VLAN) tag and a quality-of-service (QoS) data field; converting the first Layer-2 RDMA packet to a first Layer-3 encapsulated packet; and forwarding the first Layer-3 encapsulated packet to a switch fabric. In this method, the converting includes adding at least one header to the first Layer-2 RDMA packet, where the at least one header includes: a virtual network identifier that is based on information from the VLAN tag, and a QoS value that is based on information from the QoS data field.

Cloud scale multi-tenancy for RDMA over converged ethernet (RoCE)

Discussed herein is a framework that provisions for customized processing for different classes of traffic. A network device in a communication path between a source host machine and a destination host machine extracts a tag from a packet received by the network device. The packet originates at a source executing on the source host machine and whose destination is the destination host machine. The tag set by the source and indicative of a first traffic class to be associated with the packet, the first traffic class being selected by the source from a plurality of traffic classes. The network device determines, based on the tag, that the first traffic class corresponds to a latency sensitive traffic and processes the packet using one or more settings configured at the network device for processing packets associated with the first traffic class.

Network device level optimizations for latency sensitive RDMA traffic

Systems and methods for support server high availability with network link bonding for cloud overlay networks are disclosed herein. The method can include selecting a compute instance, identifying a plurality of Network Virtualization Devices ('NVD') for association with the compute instance, and creating a number of Virtualized Network Interface Cards (“VNIC”), each of which VNICs can reside in one of the plurality of NVDs. The method can include overlaying an IP address of the compute instance to each of the VNICs, such that each of the VNICs share a common IP address, designating a network path formed by one of the VNICs in one of the NVDs as an active network path and another of the network paths as an inactive network path, and activating the inactive network path when the active network path fails.

Support server high availability with network link bonding for cloud overlay networks

Techniques for controlling packet flows through the generation of packet flow rules are described. In an example, a network virtualization device receives network data. The network virtualization device determines a set of networks of a virtual network based on the network data. The network virtualization device receives flow data of the customer. The network virtualization device generates a packet flow rule based on the flow data and the set of networks. The packet flow rule defines a network boundary of one or more networks such that a first packet having a destination within the network boundary can flow and such that a second packet having a destination outside of the network boundary is to be dropped. The network virtualization device stores the packet flow rule in association with the compute instance.

Packet flow control in a header of a packet

Disclosed is an approach to implement an on-demand secure communications channel to a cloud-related resource that is located in a customer's on-premises data center, where the on-demand channel provides access to the resource to a cloud provider's operator employees. This creates on a temporary basis all of the infrastructure that is needed to allow the operational access to the customer system, which can then be destroyed once it is no longer needed.

On demand operations access to cloud customer resources

Systems and methods for highly-available host networking with active-active or active-backup traffic load-balancing are disclosed herein. The method can include selecting a compute instance from an overlay network residing on a substrate network, identifying a plurality of Network Virtualization Devices ('NVD') for association with the compute instance, creating a loopback interface on each of the NVDs, each of which loopback interfaces can include a shared IP address that can be in the substrate layer, prepopulating a table in each of the NVDs, the table linking the shared IP address to the compute instance, and each of the plurality of NVDs advertising a unique route to the compute instance via the shared IP address.

Highly-available host networking with active-active or active-backup traffic load-balancing

Systems and methods for transparent high availability for multi-customer support with hypervisor based bond implementation. The method can include creating a network path bond between a plurality of compute instances and a plurality of Network Virtualization Devices ('NVD'), the network path bond comprising a plurality of network paths, identifying a monitoring bond coupling the plurality of NVDs to a monitoring agent, creating a number of monitoring VNICs, each of the number of monitoring VNICs residing in one of the plurality of NVDs, overlaying a unique IP address to each of the monitoring VNICs, determining with the monitoring agent a health of at least one of network paths, the network paths including an active network path and an inactive network path, and activating the inactive network path when the active network path fails.

Multi-customer (multi-tenants) support with hypervisor based bond implementation

Systems and methods for transparent high availability for customer virtual machines using a hypervisor-based side channel bonding and monitoring are disclosed herein. The method can include creating a network path bond between at least one compute instance and a plurality of Network Virtualization Devices ('NVD'), the network path bond including a plurality of network paths, each network path connecting the compute instance with the Virtualized Network Interface Card ('VNIC') of one of the plurality of NVDs, identifying a first one of the network paths as an active network path and a second one of the network paths as an inactive network path, performing a health check on the active network path, determining that the active network path failed the health check, marking the first one of the network paths as failed subsequent to determining that the active network path failed the health check, and identifying the second one of the network paths as the active network path.

Transparent high availability for customer virtual machines achieved using a hypervisor-based side channel bonding and monitoring

Techniques for controlling packet flows are described. In an example, a packet is sent on a virtual network. The packet's header includes scoping data that indicates a network boundary within which the packet is permitted and/or prohibited to flow. A network virtualization device of a substrate network receives the packet. The network virtualization device determines the scoping data from the header and, based on network configuration information, determines the forward flow of the packet. If the forward flow falls within a permitted network boundary indicated by the scoping data, the network virtualization device sends the packet forward. Otherwise, the packet is dropped.

Techniques are described herein for isolating a network stack in a multitenant container DBMS. A first and second pluggable database are established within a container database. A first network stack is assigned to the first pluggable database and a second network stack is assigned to the second pluggable database. A request is received to connect to the first pluggable database and a connection is established to the first pluggable database using the first network stack. A request is received to connect to the second pluggable database and a connection is established to the second pluggable database using the second network stack.

Isolating a network stack for pluggable databases

Techniques are described herein for intelligent pooling and management of isolated hierarchical runtimes in a multi-tenant environment. In an embodiment, an allocation of operating system resources of a container database management system (CDBMS) is allocated to each generic nest of a pool of generic nests. The allocation includes a quota of one or more processors and a quota of memory. A configuration profile for a PDB in the CDBMS is then determined. Based on the configuration profile determined for the PDB, a matching generic nest from the pool of generic nests is determined. A matching generic nest is assigned to the PDB. The matching generic nest is then configured for the PDB.

Intelligent pooling of isolated hierarchical runtimes for cloud scale databases in a multi-tenant environment

Techniques are described herein for isolating runtime environments in a multitenant container DBMS. A CDB may contain a database dictionary that defines a plurality of pluggable databases and a respective database nest for each pluggable database. Each respective database nest of said each pluggable database is associated with different operating system privileges. When database sessions are established on a container DBMS, each database session is given access to a pluggable database by attaching the respective database nest of the pluggable database to each database session.

Growing community of inventors

San Jose, CA, United States of America

Santosh Narayan Shilimkar