State Exploration of Scala Actor Programs

The growing use of multicore and networked computing systems is increasing the importance of developing reliable parallel and distributed code. Unfortunately, developing and testing such code is notoriously hard, especially for shared-memory models of programming. The actor model offers a promising alternative paradigm based on message passing. Essentially, an actor is an autonomous concurrent object which interacts with other actors only by exchanging messages. In actor-based systems, the key source of non-determinism is the order in which messages are delivered to—and processed by—the actors. Bugs can still occur in actor programs as the interleaving of messages may be incorrect, or the sequential code within an actor can have bugs. We developed a general framework, called SEJAP, for exploring possible message schedules in actor systems compiled to the Java bytecode. Specifically, in this dissertation, we present one instantiation of SEJAP for the actor library of the Scala programming language. To the best of our knowledge, this is the first framework that allows systematic exploration of Scala actor programs. We also present two optimizations that alleviate the state explosion problem typical for such exploration, and thus speed up the overall exploration of actor programs in SEJAP. We have implemented our framework, Scala instantiation, and optimizations in Java PathFinder, a widely used model checker for Java bytecode developed by NASA. Preliminary results show that SEJAP can effectively explore executions of actor programs. Further, our use of SEJAP already discovered a previously unknown bug in the sample code from a popular site for Scala.

Instant Pickles: Generating Object-Oriented Pickler Combinators for Fast and Extensible Serialization

Eugene Burmako, Philipp Haller, Martin Odersky

As more applications migrate to the cloud, and as “big data” edges into even more production environments, the performance and simplicity of exchanging data between compute nodes/devices is increasing in importance. An issue central to distributed programming, yet often under-considered, is serialization or pickling, i.e., persisting runtime objects by converting them into a binary or text representation. Pickler combinators are a popular approach from functional programming; their composability alleviates some of the tedium of writing pickling code by hand, but they don’t translate well to object-oriented programming due to qualities like open class hierarchies and subtyping polymorphism. Furthermore, both functional pickler combinators and popular, Java-based serialization frameworks tend to be tied to a specific pickle format, leaving programmers with no choice of how their data is persisted. In this paper, we present object-oriented pickler combinators and a framework for generating them at compile-time, called scala/pickling, designed to be the default serialization mechanism of the Scala programming language. The static generation of OO picklers enables significant performance improvements, outperforming Java and Kryo in most of our benchmarks. In addition to high performance and the need for little to no boilerplate, our framework is extensible: using the type class pattern, users can provide both (1) custom, easily interchangeable pickle formats and (2) custom picklers, to override the default behavior of the pickling framework. In benchmarks, we compare scala/pickling with other popular industrial frameworks, and present results on time, memory usage, and size when pickling/unpickling a number of data types used in real-world, large-scale distributed applications and frameworks.

2013

Instant Pickles: Generating Object-Oriented Pickler Combinators for Fast and Extensible Serialization

Eugene Burmako, Philipp Haller, Martin Odersky

As more applications migrate to the cloud, and as "big data" edges into even more production environments, the performance and simplicity of exchanging data between compute nodes/devices is increasing in importance. An issue central to distributed programming, yet often under-considered, is serialization or pickling, i.e., persisting runtime objects by converting them into a binary or text representation. Pickler combinators are a popular approach from functional programming; their composability alleviates some of the tedium of writing pickling code by hand, but they don't translate well to object-oriented programming due to qualities like open class hierarchies and subtyping polymorphism. Furthermore, both functional pickler combinators and popular, Java-based serialization frameworks tend to be tied to a specific pickle format, leaving programmers with no choice of how their data is persisted. In this paper, we present object-oriented pickler combinators and a framework for generating them at compile-time, called scala/pickling, designed to be the default serialization mechanism of the Scala programming language. The static generation of OO picklers enables significant performance improvements, outperforming Java and Kryo in most of our benchmarks. In addition to high performance and the need for little to no boilerplate, our framework is extensible: using the type class pattern, users can provide both (1) custom, easily interchangeable pickle formats and (2) custom picklers, to override the default behavior of the pickling framework. In benchmarks, we compare scala/pickling with other popular industrial frameworks, and present results on time, memory usage, and size when pickling/unpickling a number of data types used in real-world, large-scale distributed applications and frameworks.

Assoc Computing Machinery2013

Actors that Unify Threads and Events

Philipp Haller, Martin Odersky

In practice, concurrent programming systems based on message passing are often instantiations of the actor model. A popular implementation of this form of concurrency is the Erlang programming language. Erlang supports massively concurrent systems such as telephone exchanges by providing very lightweight concurrent processes. On mainstream platforms such as the JVM, an equally attractive implementation was as yet missing. Their standard concurrency constructs, shared-memory threads with locks, suffer from high initialization and context-switching overhead as well as high memory consumption. Therefore, the interleaving of independent computations is often modelled in an event-driven style on these platforms. However, programming in an explicitly event-driven style is complicated and error-prone, because it involves an inversion of control. In this paper we present an abstraction of actors that combines the benefits of thread-based and event-based concurrency. Threads support blocking operations such as system I/O, and can be executed on multiple processor cores in parallel. Event-based computation, on the other hand, is more lightweight and scales to large numbers of actors. We also present a set of combinators that allows a flexible composition of these actors.

2007