Abstract: Technologies for adaptive network packet egress scheduling include a switch configured to configure an eligibility table for a plurality of ports of the switch, wherein the eligibility table includes a plurality of rounds. The switch is further configured to retrieve an eligible mask corresponding to a round of a plurality of rounds of the eligibility table presently being scheduled and determine a ready mask that indicates a ready status of each port. The switch is further configured to determine, for each port, whether the eligible status and the ready status indicate that port is both eligible and ready, and schedule, in response to a determination that at least one port has been determined to be both eligible and ready, each of the at least one port that has been determined to be both eligible and ready. Additional embodiments are described herein.

Abstract: Technologies for updating an access control list (ACL) table while minimizing disruption includes a network device. The network device receives a request to store a rule in the ACL. The rule is associated with a precedence group. A precedence group is indicative of a placement priority of a given rule in the ACL. The network device determines, as a function of the precedence group, a placement for the requested rule in the ACL. The network device stores the rule according to the determined placement in the ACL.

Abstract: There is disclosed in one example a network switching apparatus, including: a plurality ingress port; a plurality of egress ports; a ternary content addressable memory (TCAM) comprising a plurality of chunks, wherein the chunks can be atomically enabled or disabled; a switching circuit to switch traffic from the ingress port to a selected egress port according to an access control list (ACL) of the TCAM; and one or more non-transitory mediums having stored thereon instructions to atomically add or update two or more target rules, including: add the two or more target rules to one or more target-rule chunks; and atomically enable the target-rule chunks.

Abstract: Examples include a computing system having a load balancer circuitry to assign data units to destinations using a hash function having a plurality of hash bins, each hash bin being associated with a destination. The computing system includes a hash adjuster to assign a destination to each of the plurality of hash bins, sample assignment of data units to hash bins by the load balancer circuitry over a sample window, analyze the sampled assignments, and reassign destinations to the hash bins based at least in part on the analyzed sampled assignments. This results in increased system performance in processing of the data units by the load balancer.

Abstract: Examples include a computing system having a load balancer circuitry to assign data units to destinations using a hash function having a plurality of hash bins, each hash bin being associated with a destination. The computing system includes a hash adjuster to assign a destination to each of the plurality of hash bins, sample assignment of data units to hash bins by the load balancer circuitry over a sample window, analyze the sampled assignments, and reassign destinations to the hash bins based at least in part on the analyzed sampled assignments. This results in increased system performance in processing of the data units by the load balancer.

Abstract: Technologies for applying a scrambling function in unfair load balancing scenarios include a network device having a communication circuitry and compute engine executing software on the network device. The compute engine is to determine, using a load balancing function, a distribution for data sets to each of multiple bins. Each bin maps to a destination. The network device determines, as a function of the distribution, an unfairness score indicative of the distribution being unequal among each of the destinations. In response to a determination that the unfairness score exceeds a specified threshold, the network device enables a scrambling function to be performed during configuration of the plurality of bins with the destinations.

Abstract: Technologies for adaptive network packet egress scheduling include a switch configured to configure an eligibility table for a plurality of ports of the switch, wherein the eligibility table includes a plurality of rounds. The switch is further configured to retrieve an eligible mask corresponding to a round of a plurality of rounds of the eligibility table presently being scheduled and determine a ready mask that indicates a ready status of each port. The switch is further configured to determine, for each port, whether the eligible status and the ready status indicate that port is both eligible and ready, and schedule, in response to a determination that at least one port has been determined to be both eligible and ready, each of the at least one port that has been determined to be both eligible and ready. Additional embodiments are described herein.

Abstract: There is disclosed in one example a network switching apparatus, including: a plurality ingress port; a plurality of egress ports; a ternary content addressable memory (TCAM) comprising a plurality of chunks, wherein the chunks can be atomically enabled or disabled; a switching circuit to switch traffic from the ingress port to a selected egress port according to an access control list (ACL) of the TCAM; and one or more non-transitory mediums having stored thereon instructions to atomically add or update two or more target rules, including: add the two or more target rules to one or more target-rule chunks; and atomically enable the target-rule chunks.

Abstract: Technologies for applying a scrambling function in unfair load balancing scenarios include a network device having a communication circuitry and compute engine executing software on the network device. The compute engine is to determine, using a load balancing function, a distribution for data sets to each of multiple bins. Each bin maps to a destination. The network device determines, as a function of the distribution, an unfairness score indicative of the distribution being unequal among each of the destinations. In response to a determination that the unfairness score exceeds a specified threshold, the network device enables a scrambling function to be performed during configuration of the plurality of bins with the destinations.

Abstract: Technologies for updating an access control list (ACL) table while minimizing disruption includes a network device. The network device receives a request to store a rule in the ACL. The rule is associated with a precedence group. A precedence group is indicative of a placement priority of a given rule in the ACL. The network device determines, as a function of the precedence group, a placement for the requested rule in the ACL. The network device stores the rule according to the determined placement in the ACL.