Browse Prior Art Database

Metaparallelism - Branch Prediction Mechanism

IP.com Disclosure Number: IPCOM000105709D
Original Publication Date: 1993-Sep-01
Included in the Prior Art Database: 2005-Mar-20
Document File: 4 page(s) / 147K

Publishing Venue

IBM

Related People

Ekanadham, K: AUTHOR [+2]

Abstract

Metaparallelism is a process that determines the form of parallelism that is to be used in a specific application. Metaparallelism has two interfaces:

This text was extracted from an ASCII text file.
This is the abbreviated version, containing approximately 33% of the total text.

Metaparallelism - Branch Prediction Mechanism

      Metaparallelism is a process that determines the form of
parallelism that is to be used in a specific application.
Metaparallelism has two interfaces:

o   Information derived from prior executions.
o   Explicit statements made in the program or compiler output that
    bear on the form of parallelism.

Metaparallelism uses aspects of program behavior as it relates to the
capabilities of the Metaparallel Processor to cope with this behavior
to determine the type of parallelism that is to be pursued.
Metaparallelism employes speculation, that is, allocation of
resources to computations without a guarantee that these computations
are required, in order to complete the application in the faster
time.

The choice of parallelisms that metaparallelism can select from are:

o   Path-oriented forms of parallelism.
o   Path-oriented forms of parallelism with speculation.
o   Computation-oriented parallelism with bifurcation at branches.
o   A set of independent paths that intercommunicate by sending
    messages to each other.

o   A combination of the above.

Metaparallelism employs means at its disposal to alter the form of
parallelism specified by the programmer/compiler at the source level
and to notify the programmer about significant aspects that interfere
with the parallelization of the application.

      The interrelations are between the instructions that, set
values of registers and the instructions that use those values set.
This interrelationship can be formulated in terms of a Directed
Acyclic Graph (DAG).  The nodes of the DAG are instructions, and the
arcs represent interrelationships of set/use.

      The components of the conditional DAG are DAG for instruction
interrelationship that relate to computations.  A computation is a
set of sequential instructions that are between two variable
action/target, TYPE-B, branches.  The DAG for the i computation is
conditional in the sense that the computation being only one of
several targets of a variable branch may not be part of the path.  If
it is part of the path, then this DAG can be simply merged with the
DAGs for the other parts of the DAG to create the overall DAG.  The
DAG for a computation has three components that are interrelated:

o   Place-Holder-On-Entry.

          This allows instructions within the computation to show how
    they depend on values of the register set prior to the
    computation and which register is not modified by the
    computation.

o   DAG between instructions within the computation.

          This shows which instructions within the computation set
    inputs for other instructions in the computation.

o   Place-Holder-On-Exit.

          This allows instructions within the computation to show how
    they set values of the registers that may be used by computations
    that follow this computation and which registers is not...