Meg Walraed-Sullivan

Aspen trees: balancing data center fault tolerance, scalability and cost

Fault recovery is a key issue in modern data centers. In a fat tree topology, a single link failure can disconnect a set of end hosts from the rest of the network until updated routing information is disseminated to every switch in the topology. The time for re-convergence can be substantial, leaving hosts disconnected for long periods of time and significantly reducing the overall availability of the data center. Moreover, the message overhead of sending updated routing information to the entire topology may be unacceptable at scale. We present techniques to modify hierarchical data center topologies to enable switches to react to failures locally, thus reducing both the convergence time and control overhead of failure recovery. We find that for a given network size, decreasing a topologys convergence time results in a proportional decrease to its scalability (e.g. the number of hosts supported). On the other hand, reducing convergence time without affecting scalability necessitates the introduction of additional switches and links. We explore the tradeoffs between fault tolerance, scalability and network size, and propose a range of modified multi-rooted tree topologies that provide significantly reduced convergence time while retaining most of the traditional fat trees desirable properties.

ALIAS: scalable, decentralized label assignment for data centers

Modern data centers can consist of hundreds of thousands of servers and millions of virtualized end hosts. Managing address assignment while simultaneously enabling scalable communication is a challenge in such an environment. We present ALIAS, an addressing and communication protocol that automates topology discovery and address assignment for the hierarchical topologies that underlie many data center network fabrics. Addresses assigned by ALIAS interoperate with a variety of scalable communication techniques. ALIAS is fully decentralized, scales to large network sizes, and dynamically recovers from arbitrary failures, without requiring modifications to hosts or to commodity switch hardware. We demonstrate through simulation that ALIAS quickly and correctly configures networks that support up to hundreds of thousands of hosts, even in the face of failures and erroneous cabling, and we show that ALIAS is a practical solution for auto-configuration with our NetFPGA testbed implementation.

Theia: Simple and Cheap Networking for Ultra-Dense Data Centers

Recent trends to pack data centers with more CPUs per rack have led to a scenario in which each individual rack may contain hundreds, or even thousands, of compute nodes using system-on-chip (SoC) architectures. At this increased scale, traditional rack-level star topologies with a top-of-rack (ToR) switch as the hub and servers as the leaves are no longer feasible in terms of monetary cost, physical space, and oversubscription. We propose Theia, an architecture to connect hundreds of SoC nodes within a rack, using inexpensive, low-latency, hardware elements to group the racks servers into subsets which we term SubRacks. We then replace the traditional per-rack ToR with a low-latency, passive, circuit-style patch panel that interconnects these SubRacks. We explore alternatives for the rack-level topology implemented by this patch panel, and we consider approaches for interconnecting racks within a data center. Finally, we investigate options for routing over these new topologies. Our proposal of Theia is unique in that it offers the flexibility of a packet-switched networking over a fixed circuit topology.

Fast encounter-based synchronization for mobile devices

The large and growing number of computing devices used by individuals has caused the challenges of distributed storage to take on increased importance. In addition to desktops and laptops, portable devices, such as cell phones, digital cameras, iPods, and PDAs are capable of storing and sharing data. These devices’ mobility coupled with wireless networking capabilities allows them to opportunistically propagate data to other devices they might encounter, even if connectivity is unplanned and transient. To take advantage of such ad hoc connectivity, devices must have an efficient method for determining which files they hold in common and which versions must be propagated to achieve consistency. This paper presents new techniques for reducing the cost of this data synchronization operation by over an order of magnitude in many situations.

Brief announcement: a randomized algorithm for label assignment in dynamic networks

In large-scale networking environments, such as data centers, a key difficulty is the assignment of labels to network elements. Labels can be assigned statically, e.g. MAC-addresses in traditional Layer 2 networks, or by a central authority as in DHCP in Layer 3 networks. On the other hand, networks requiring a dynamic solution often use a Consensus-based state machine approach. While designing Alias [2], a protocol for automatically assigning hierarchically meaningful addresses in data center networks, we encountered an instance of label assignment with entirely different requirements. In this case, the rules for labels depend on connectivity, and connectivity (and hence, labels) changes over time. Thus, neither static assignment nor a state machine approach is ideal.