skip to main content


Title: TEA: Enabling State-Intensive Network Functions on Programmable Switches
Programmable switches have been touted as an attractive alternative for deploying network functions (NFs) such as network address translators (NATs), load balancers, and firewalls. However, their limited memory capacity has been a major stumbling block that has stymied their adoption for supporting state-intensive NFs such as cloud-scale NATs and load balancers that maintain millions of flow-table entries. In this paper, we explore a new approach that leverages DRAM on servers available in typical NFV clusters. Our new system architecture, called TEA (Table Extension Architecture), provides a virtual table abstraction that allows NFs on programmable switches to look up large virtual tables built on external DRAM. Our approach enables switch ASICs to access external DRAM purely in the data plane without involving CPUs on servers. We address key design and implementation challenges in realizing this idea. We demonstrate its feasibility and practicality with our implementation on a Tofino-based programmable switch. Our evaluation shows that NFs built with TEA can look up table entries on external DRAM with low and predictable latency (1.8-2.2 μs) and the lookup throughput can be linearly scaled with additional servers (138 million lookups per seconds with 8 servers).  more » « less
Award ID(s):
1700521
NSF-PAR ID:
10180448
Author(s) / Creator(s):
; ; ; ; ; ;
Date Published:
Journal Name:
SIGCOMM '20: Proceedings of the Annual conference of the ACM Special Interest Group on Data Communication on the applications, technologies, architectures, and protocols for computer communication
Page Range / eLocation ID:
90 to 106
Format(s):
Medium: X
Sponsoring Org:
National Science Foundation
More Like this
  1. null (Ed.)
    Many recent efforts have demonstrated the performance benefits of running datacenter functions (e.g., NATs, load balancers, monitoring) on programmable switches. However, a key missing piece remains: fault tolerance. This is especially critical as the network is no longer stateless and pure endpoint recovery does not suffice. In this paper, we design and implement RedPlane, a fault-tolerant state store for stateful in-switch applications. This provides in-switch applications consistent access to their state, even if the switch they run on fails or traffic is rerouted to an alternative switch. We address key challenges in devising a practical, provably correct replication protocol and implementing it in the switch data plane. Our evaluations show that RedPlane incurs negligible overhead and enables end-to-end applications to rapidly recover from switch failures. 
    more » « less
  2. Modern programmable network switches can implement cus- tom applications using efficient packet processing hardware, and the programming language P4 provides high-level con- structs to program such switches. The increase in speed and programmability has inspired research in dataplane program- ming, where many complex functionalities, e.g., key-value stores and load balancers, can be implemented entirely in network switches. However, dataplane programs may suffer from novel security errors that are not traditionally found in network switches. To address this issue, we present a new information-flow control type system for P4. We formalize our type system in a recently-proposed core version of P4, and we prove a sound- ness theorem: well-typed programs satisfy non-interference. We also implement our type system in a tool, P4BID, which extends the type checker in the p4c compiler, the reference compiler for the latest version of P4. We present several case studies showing that natural security, integrity, and isolation properties in networks can be captured by non-interference, and our type system can detect violations of these properties while certifying correct programs. 
    more » « less
  3. Traditional network resident functions (e.g., firewalls, network address translation) and middleboxes (caches, load balancers) have moved from purpose-built appliances to software-based components. However, L2/L3 network functions (NFs) are being implemented on Network Function Virtualization (NFV) platforms that extensively exploit kernel-bypass technology. They often use DPDK for zero-copy delivery and high performance. On the other hand, L4/L7 middleboxes, which usually require full network protocol stack support, take advantage of a full-fledged kernel-based system with a greater emphasis on functionality. Thus, L2/L3 NFs and middleboxes continue to be handled by distinct platforms on different nodes.This paper proposes MiddleNet that seeks to overcome this dichotomy by developing a unified network resident function framework that supports L2/L3 NFs and L4/L7 middleboxes. MiddleNet supports function chains that are essential in both NFV and middlebox environments. MiddleNet uses DPDK for zero-copy packet delivery without interrupt-based processing, to enable the ‘bump-in-the-wire’ L2/L3 processing performance required of NFV. To support L4/L7 middlebox functionality, MiddleNet utilizes a consolidated, kernel-based protocol stack processing, avoiding a dedicated protocol stack for each function. MiddleNet fully exploits the event-driven capabilities provided by the extended Berkeley Packet Filter (eBPF) and seamlessly integrates it with shared memory for high-performance communication in L4/L7 middlebox function chains. The overheads for MiddleNet are strictly load-proportional, without needing the dedicated CPU cores of DPDK-based approaches. MiddleNet supports flow-dependent packet processing by leveraging Single Root I/O Virtualization (SR-IOV) to dynamically select packet processing needed (Layer 2 to Layer 7). Our experimental results show that MiddleNet can achieve high performance in such a unified environment. 
    more » « less
  4. null (Ed.)
    This paper considers the MapReduce-like coded distributed computing framework originally proposed by Li et al., which uses coding techniques when distributed computing servers exchange their computed intermediate values, in order to reduce the overall traffic load. In their original model, servers are connected via an error-free common communication bus allowing broadcast transmissions. However, this assumption is one of the major limitations for practical implementations since real-world data centers may have network topologies far more involved than a single broadcast bus. We formulate a topological coded distributed computing problem, where the computing servers communicate with each other through some switch network. By using a special instance of fat-tree topologies, referred to as t-ary fat-tree proposed by Al-Fares et al. which can be built by some inexpensive switches, we propose a coded distributed computing scheme to achieve the optimal max-link communication load (defined as the maximum load over all links) over any network topology. 
    more » « less
  5. Intrusion detection through classifying incoming packets is a crucial functionality at the network edge, requiring accuracy, efficiency and scalability at the same time, introducing a great challenge. On the one hand, traditional table-based switch functions have limited capacity to identify complicated network attack behaviors. On the other hand, machine learning based methods providing high accuracy are widely used for packet classification, but they typically require packets to be forwarded to an extra host and therefore increase the network latency. To overcome these limitations, in this paper we propose an architecture with programmable data plane switches. We show that Binarized Neural Networks (BNNs) can be implemented as switch functions at the network edge classifying incoming packets at the line speed of the switches. To train BNNs in a scalable manner, we adopt a federated learning approach that keeps the communication overheads of training small even for scenarios involving many edge network domains. We next develop a prototype using the P4 language and perform evaluations. The results demonstrate that a multi-fold improvement in latency and communication overheads can be achieved compared to state-of the-art learning architectures. 
    more » « less