A first plurality of UDP packets corresponding to a first session and a second plurality of UDP packets corresponding to a second session can be received at a core network node. The first plurality of UDP packets and the second plurality of UDP packets have the same source five tuple. A first plurality of HTTP packets comprising first session information identifying the first session can be created based on the first plurality of UDP packets. A second plurality of HTTP packets comprising second session information identifying the second session can be created based on the second plurality of UDP packets. The first plurality of HTTP packets can be routed to a first microservice instance based at least in part on the first session information. The second plurality of HTTP packets can be routed to a second microservice instance based at least in part on the second session information.

Systems and methods for establishing a multipath connection include a first processor of a first cluster forwarding a first request from a client to establish a first connection with a server to a second processor of a second cluster. A third processor of the first cluster receives a second request to establish a multipath connection between the client and the server. The third processor forwards the second request to the second processor responsive to determining that the second request is to establish a multipath connection. The second processor establishes the multipath connection that includes the first connection and a second connection used as paths of the multipath connection.

In some examples, a system receives a first collection of tokens relating to characteristics of workloads for a computing system. The system encodes the first collection of tokens, the encoding including computing weights representing relationships among tokens of the first collection of tokens, and generating a representation of the first collection of tokens based on the weights. The system determines, based on the representation, a correlation between the first collection of tokens and a second collection of tokens relating to elapsed times in executing the workloads.

In some examples, a system receives a first collection of tokens relating to characteristics of workloads for a computing system. The system encodes the first collection of tokens, the encoding including computing weights representing relationships among tokens of the first collection of tokens, and generating a representation of the first collection of tokens based on the weights. The system determines, based on the representation, a correlation between the first collection of tokens and a second collection of tokens relating to elapsed times in executing the workloads.

In accordance with an embodiment, described herein is a system and method for supporting an object-based stateful protocol (for example, T3 protocol) in a multitenant application server environment. The system includes a domain with a plurality of server instances, and a plurality of partitions thereon, wherein two or more partitions can spread across a first server instance and a second server instance of the plurality of server instances. Each of the first and second server instances can include a single remote java virtual machine (RJVM) that manages connections established through a single server channel on that server instance using the object-based stateful protocol. The application server instance initiating the connections can associate each connection with a particular partition using a token, and use a smart stub to recover a failed connection. A standalone Java application can similarly initiate connections to an application server instance.

In one embodiment, a system uses a two-layer deterministic function, such as a hashing mechanism, to ensure session stickiness or affinity. A client can generate a session key that can be used for at least two things. First, a load balancer can be selected by the client using the session key by using a consistent or deterministic hashing algorithm so that the same load balancer is selected from a set of weighted load balancers. After selecting the load balancer, the client can send the request to the selected load balancer. The load balancer then uses the same session key, which can be transmitted in an HTTP header, to select a backend host server computer. The result is that session affinity is maintained.

Embodiments are directed towards managing persistence of network traffic using deep packet inspections of network response packets from an application server. In one embodiment, the network packets are associated with SIP messages. A traffic management device (TMD) interposed between client devices and a plurality of application servers receives messages from the client device and/or the application servers. The TMD performs a deep packet inspection to determine if a defined key value pair that includes a session identifier is detected. If so, and the message is from the application server, the session identifier is then mapped to an application server identifier to persistently refer each subsequent inbound packet from a client device having the same session identifier to the application server mapped to the session identifier.

Dynamic advertisement routing is disclosed. For example, a plurality of internet protocol (“IP”) addresses associated with respective plurality of target nodes is stored in a routing pool. Each IP address in the routing pool is pinged through each of first and second load balancer network interfaces. Network routes associated with target nodes are updated based on a first plurality of ping responses. Communications sessions are established with target nodes through respective network routes. IP addresses are pinged and respective latencies in a latency cache are updated based on a second plurality of ping responses. A first request directed to the plurality of target nodes is received and is determined to be sent to a first target node based on the latency cache forwarded to the first target node via the first network route.

A non-transitory computer readable medium comprising instructions which causes performance of operations comprising: receiving, by a second network coordination device, current state information of a client device, including a transmit counter and a receive counter, from a first network coordination device, wherein the first network coordination device is a primary network coordination device for the client device and the second network coordination device is a standby coordination device for the client device; and responsive to detecting a particular event, the second network coordination device (i) transitioning to be the primary network coordination device for the client device, (ii) incrementing the transmit counter by an offset, and (iii) transmitting a message to the client device including the incremented transmit counter is shown.

A method for facilitating session data persistence and management is disclosed. The method includes receiving a first indication from an application, the first indication relating to an initiation of a first session; compiling session data from the application, the session data relating to a state of the application; receiving a termination indication from the application, the termination indication relating to a termination of the first session; persisting, in a session cache, the session data based on the termination indication; receiving a second indication from the application, the second indication relating to an initiation of a second session; identifying the corresponding session data in the session cache by using the second indication; and automatically injecting the identified session data into the application.