PECASE: OceanStore: An Oceanic Data Utility for Ubiquitous, Highly-Available, Reliable, and Persistant Storage

University Of California-Berkeley, Berkeley CA

Investigators

Abstract

This is an NSF CAREER/PECASE grant proposal for research into Extremely Wide-Area StorageSystems (EWASS). Ideally, such storage systems are highly-available from anywhere in the network, exploit automatic replication for disaster recovery, employ strong security by default, and provide performance that is similar to that of local storage under numerous circumstances. Further, such systems are self-repairing, providing automatic serviceability and maintainability. The research envisions a utility model in which users pay a monthly fee to one particular "utility provider" while consuming resources from many providers. The utility providers buy and sell capacity amongst themselves in direct analogy to the deregulated electric industry. A demand for capacity or bandwidth in one region of the country would encourage entrepreneurs to bring additional services online. One of the key requirements for such utilities is nomadic data, i.e., data that is free to migrate and be replicated anywhere within the system. Nomadic data lends itself to fluid analogies (data is free to "flow" wherever it is needed) and permits flexible exploitation of spatial locality. This property is in contrast to existing distributed systems that confine their data to a small number of physical servers within the network. Given the fluid analogy, the EWASS prototype described within this proposal is called the "Oceanic Data Utility" or "OceanStore". The project outlines five assumptions for the OceanStore utility: (1) The "Mostly Well-Connected" Assumption: Most of the network is comprised of high-bandwidth links and periods of disconnection are brief - even at the leaves. (2) The "Promiscuous Caching" Assumption: Data can be cached anywhere, anytime. (3)The "Operations-based Interface with Conflict Resolution" Assumption: Applications utilize an operation-based interface that is oriented toward conflict resolution. (4)The "Untrusted Infrastructure" Assumption: The infrastructure is fundamentally untrusted. This means that only ciphertext is stored in the network. (5)The "Responsible Party" Assumption: Each repository of data has at least one responsible party that is charged with knowing where the data actually resides and ensuring that it is properly replicated. These assumptions define the OceanStore infrastructure. In constructing this infrastructure, the research proposes to utilize a number of technologies, some of them mature, others relatively new: (1)Organization of data as a series of "pools" of information. Each pool will include a randomized tree indexing structures (e.g., treaps) connected via Bloom-filter summaries. (2)Transaction-like operation structures to describe updates and permit infrastructure-side conflict resolution (similar to Bayou), combined with incremental cryptographic techniques to operate directly on encrypted information in the infrastructure. (3)Use of erasure-tolerant coding and replication to enhance the survivability of data. (4) Online Introspection to detect and exploit patterns of access in order to optimize the position of nomadic data. The research proposes two distinct phases of OceanStore prototype, one read-only, the other complete with conflict resolution. This phased style of implementation permits incremental testing and analysis of components of the system that are somewhat orthogonal. Further, we propose to use the OceanStore infrastructure to enhance education at Berkeley in several ways. Among them: (1) Collaborate with other faculty to combine the persistent storage system with experimental wireless infrastructures in order to exploit online content during class and to provide novel faculty/student interactions at nonstandard venues such as cafes. (2) Provide a fundamental infrastructure for undergraduates and graduates exploiting the consequences of ubiquitous computing. OceanStore provides a foundation for ubiquitous computing, and as such will doubtless have a major impact on educational paradigms.