Protium, an infrastructure for partitioned applications

Abstract

Remote access feels different from local access. The major issues are consistency (machines vary in GUls, applications, and devices) and responsiveness (the user must wait for network and server delays). Protium attacks these by partitioning programs into local viewers that connect to remote services using application-specific protocols. Partitioning allows viewers to be customized to adapt to local features and limitations. Services are responsible for maintaining long-term state. Viewers manage the user interface and use state to reduce communication between viewer and service, reducing latency whenever possible.System infrastructure sits between the viewer and service, supporting replication, consistency, session management, and multiple simultaneous viewers. The prototype system includes an editor, a draw program, a PDF viewer, a map database, a music jukebox, and windowing system support. It runs on servers, workstations, PCs, and PDAs under Plan 9, Linux, and Windows; services and viewers have been written in C, Java, and Concurrent ML.

Ignis: scaling distribution-oblivious systems with light-touch distribution

“…Packages and Modules Modules provide an implicit and fine-grained component architecture [26,27,35,62] that applications can be partitioned across [18,37,63,74,91]. They encapsulate state behind small and tight interfaces, simplifying and minimizing transformations.…”

Energy optimizations for mobile terminals via computation offloading

“…Protium [7] manually splits program code into viewers and services, one runs near the user, while the other runs in a service provider that is highly available, with persistent storage and abundant computation cycles. Viewers and services communicate via an applicationspecific protocol.…”

“…Single server Static Manually Coarse-grained [7] Data center Static Manually Fine-grained [8] Crowd Static Automatic Fine-grained [9] N/A Static Automatic Fine-grained [10] Single server Static Automatic Fine-grained [11] N/A Dynamic Automatic Fine-grained [12] N/A Dynamic Automatic Fine-grained [13] Data center Dynamic Automatic Fine-grained [14] Data center Dynamic Automatic Fine-grained [15] N/A Static Automatic Coarse-grained [16] N/A Static Automatic Coarse-grained [17] Data center Dynamic Automatic Coarse-grained [18] Data center Static Automatic Coarse-grained [3] [8] [9] [10] all belong to this scheme. The communication cost depends on the size of the transmitted data and the network bandwidth, and the computation cost depends on the number of program instructions.…”

Programming in Manticore, a Heterogenous Parallel Functional Language

“…This mechanism allows programmers to encapsulate complicated communication and synchronization protocols as first-class abstractions, which encourages a modular style of programming where the actual underlying channels used to communicate with a given thread are hidden behind data and type abstraction. CML has been used successfully in a number of systems, including a multithreaded GUI toolkit [34], a distributed tuple-space implementation [71], and a system for implementing partitioned applications in a distributed setting [81]. The design of CML has inspired many implementations of CML-style concurrency primitives in other languages.…”