Processors that adapt architecture to workloads at runtime promise compelling performance per watt (PPW) gains, offering one way to mitigate diminishing returns from pipeline scaling. State-of-the-art adaptive CPUs deploy machine learning (ML) models on-chip to optimize hardware by recognizing workload patterns in event counter data. However, despite breakthrough PPW gains, such designs are not yet widely adopted due to the potential for systematic adaptation errors in the field.

We introduce State-Informed Link-Layer Queuing (SILQ), a system that models, predicts, and avoids packet delivery failures caused by temporary wireless outages in everyday scenarios. By stabilizing connections in adverse link conditions, SILQ boosts throughput and reduces performance variation for network applications, for example by preventing unnecessary TCP timeouts due to dead zones, elevators, and subway tunnels. SILQ makes predictions in real-time by actively probing links, matching measurements to an overcomplete dictionary of patterns learned offline, and classifying the resulting sparse feature vectors to identify those that precede outages. We use a clustering method called sparse coding to build our data-driven link model, and show that it produces more variation-tolerant predictions than traditional loss-rate, location-based, or Markov chain techniques.

Wireless data transfer under high mobility, as found in unmanned aerial vehicle (UAV) applications, is a challenge due to varying channel quality and extended link outages. We present FlowCode, an easily deployable link-layer solution utilizing multiple transmitters and receivers for the purpose of supporting existing transport protocols such as TCP in these scenarios. By using multiple transmitters and receivers and by exploiting the resulting antenna beam diversity and parallel transmission effects, FlowCode increases throughput and reception range. In emulation, we show that TCP over FlowCode gives greater goodput over a larger portion of the flight path, compared to an enhanced TCP protocol using the standard 802.11 MAC. In the process, we make a strong case for using trace-modulated emulation when developing distributed protocols for complex wireless environments.

Firewalls inspect arriving packets according to a security policy. The complexity of these policies can cause significant delays in the processing of packets, resulting in degraded performance, traffic bottlenecks, and ultimately violating Quality of Service (QoS) constraints. As network capacities continue to increase, the improvement of firewall performance is a main concern. One technique that dramatically reduces required processing is the representation of security policies in software with n-ary tries. This paper describes trie balancing methods that further improve performance by placing more frequently used rules in high precedence positions which require fewer tuple comparisons. A proof of sorted trie integrity is presented along with experimental results showing that on average, sorting reduces the number of comparisons by 27% as compared to the original trie and by 83% as compared to a list representation. Sorting methods are described in detail and their benefits are demonstrated empirically.

Network firewalls remain the forefront defense for most computer systems. These critical devices filter traffic by comparing arriving packets to a list of rules, or security policy, in a sequential manner. Unfortunately packet filtering in this fashion can result in significant traffic delays, which is problematic for applications that require strict quality of service (QoS) guarantees. Given this demanding environment, new methods are needed to increase network firewall performance. This paper introduces a new technique for representing a security policy that maintains policy integrity and provides more efficient processing. The policy is represented as an n-ary retrieval tree, also referred to as a trie. The worst case processing requirement for the policy trie is a fraction compared a list representation, which only considers rules individually (1/5 the processing for TCP/IP networks). Furthermore unlike other representations, the n-ary trie developed in this paper can be proven to maintain policy integrity. The creation of policy trie structures is discussed in detail and their performance benefits are described theoretically and validated empirically.