This model is implemented as a "coordination language" in which several primitives operating on ordered sequence of typed data objects, "tuples," are added to a sequential language, such as C, and a logically global associative memory, called a tuplespace, in which processes store and retrieve tuples. The original Linda model requires four operations that individual workers perform on the tuples and the tuplespace:
in atomically reads and removes—consumes—a tuple from tuplespace
rd non-destructively reads a tuplespace
out produces a tuple, writing it into tuplespace
eval creates new processes to evaluate tuples, writing the result into tuplespace
Linda-calculus
The Linda-calculus is a formalisation of the above model with the difference that in the following subsumes both out and eval operations. The presentation of the calculus is based on Cridlig et al.
Syntax
We abstract the concrete representation of tuples. We just assume that we have a set of tuples and we are allowed to form and apply a substitution function on tuples substituting variables for terms that yields a tuple. For example, given we have a tuple, then applying a substitution on yields The Linda-calculus processes are defined by the following grammar. The syntax includes the aftermentioned Linda operations, non-deterministic choice, and recursion. The substitution function is extended to processes recursively.
Semantics
A tuple space is represented as a multiset of the processes. We write for where is a multiset, a singleton multiset, and is the multiset union operation. The semantics is then defined as a reduction relation on a multiset as follows. Note that consumes the tuple from the tuple space whereas only reads it. The resulting operational semantics is synchronous.
Evaluation
Compared to other parallel-processing models, Linda is more orthogonal in treating process coordination as a separate activity from computation, and it is more general in being able to subsume various levels of concurrency—uniprocessor, multi-threaded multiprocessor, or networked—under a single model. Its orthogonality allows processes computing in different languages and platforms to interoperate using the same primitives. Its generality allows a multi-threaded Linda system to be distributed across multiple computers without change. Whereas message-passing models require tightly-coupled processes sending messages to each other in some sequence or protocol, Linda processes are decoupled from other processes, communicating only through the tuplespace; a process need have no notion of other processes except for the kinds of tuples consumed or produced. Researchers have proposed more primitives to support different types of communication and co-ordination between computer systems, and to solve particular problems arising from various uses of the model. Researchers have also experimented with various means of implementing the virtual shared memory for this model. Many of these researchers proposed larger modifications to the original Linda model, developing a family of systems known as Linda-like systems and implemented as orthogonal technology. An example of this is the languageEase designed by Steven Ericsson-Zenith.
Implementations
Linda was originally implemented in C and Fortran, but has since been implemented in many programming languages, including:
JavaSpaces - a Java-based tuplespace implementation that helped popularize distributed computing.
- a Java-based tuplespace platform from IBM.
Criticisms
Criticisms of Linda from the multiprocessing community tend to focus on the decreased speed of operations in Linda systems as compared to Message Passing Interface systems. While not without justification, these claims were largely refuted for an important class of problems. Detailed criticisms of the Linda model can also be found in Steven Ericsson-Zenith's book Process Interaction Models.
