Author Topic: Implementation of Embedded Multiprocessor Architecture Using FPGA  (Read 4692 times)

0 Members and 1 Guest are viewing this topic.

IJSER Content Writer

  • Sr. Member
  • ****
  • Posts: 327
  • Karma: +0/-1
    • View Profile
Quote
Author : Mr.Sumedh.S.Jadhav, Prof.C.N.Bhoyar
International Journal of Scientific & Engineering Research Volume 3, Issue 1, January-2012
ISSN 2229-5518
Download Full Paper : PDF

Abstract—   Embedded multiprocessor design presents challenges and opportunities that stem from task coarse granularity and the large number of inputs and outputs for each task. We have therefore designed a new architecture called embedded concurrent computing (ECC), which is implementing on FPGA chip using VHDL. The design methodology is expected to allow scalable embedded multiprocessors for system expansion. In recent decades, two forces have driven the increase of the processor performance: Advances in very large-scale integration (VLSI) technology and Micro architectural enhancements. Therefore, we aim to design the full architecture of an embedded processor for realistic to perform arithmetic, logical, shifting and branching operations. We will be synthesize and evaluated the embedded system based on Xilinx environment. Processor performance is going to be improving through clock speed increases and the clock speed increases and the exploitation of instruction- level parallelism. We will be designing embedded multiprocessor based on Xilinx environment. These data have been gathered by synthesis. Implementation was achieved by the provision of low complexities in terms of FPGA resource usage and frequency. In addition, the design methodology allows scalable embedded multiprocessors for system expansion.
 
Index Terms— Embedded System design, FPGA ,Memory Architecture, Real time Processor, Scheduler, VHDL Environment, Xilinx Environment.
 
1 INTRODUCTION
In  recent decades, two forces have driven the increase of the processor performance: Firstly, advances in very large-scale integration (VLSI) technology and secondly micro architectural enhancements [1].
     Processor Performance has been improve  through clock speed Increases and the exploitation of instruction-level Parallelism. While transistor counts continue to increase, recent attempts to achieve even more significant increase in single-core performance have brought diminishing returns [2, 3]. In response, architects are building chips With multiple energy-efficient processing cores instead of investing the whole transistor count into a single, complex, and power-inefficient core [3, 4]. Modern embedded systems are design as systems-on a-chip (SoC) that incorporate single chip multiple Programmable cores ranging from single chip multiple programmable cores ranging from processors to custom designed accelerators. This paradigm allows the reuse of pre-designed cores, simplifying the design of billion transistor chips, and amortizing costs.

In the past few years, parallel-programmable SoC (PPSoC)have Successful PPSoC are high-performance embedded multiprocessors such as the STI Cell [3] .They are dubbed single-chip heterogeneous multiprocessors (SCHMs) because they have a dedicated processor that coordinates the rest of the processing units. A multiprocessor design with SoC like integration of less-efficient, general-purpose processor cores with more efficient special-purpose helper engines is project to be the next step in computer evolution [5].
    First, we aim to design the full architecture of an embedded processor for realistic throughput. We used FPGA technology not only for architectural exploration
but also as our target deployment platform because we believe that this approach is best for validating the feasibility of an efficient hardware implementation.
     This architecture of the embedded processor resembles a superscalar pipeline, including the fetch, decode, rename, and dispatch units as parts of the in-order front-end. The out of-order execution core contains the task queue, dynamic scheduler; execute unit, and physical register file. The in order back-end is comprised of only the retire unit. The embedded architecture will be implement using the help of RTL descriptions in System VHDL.
    We will integrate the embedded processor with a shared memory system, synthesized this system on an FPGA environment, and performed several experiments using realistic benchmarks. the methodology to design and implement a microprocessor or multiprocessors is presented. To illustrate it with high detail and in a useful way, how to design the most complex practical session is shown. In most cases, computer architecture has been taught with software simulators [1], [2]. These simulators are useful to show: internal values in registers, memory accesses, cache fails, etc. However, the structure of the microprocessor is not visible.
    In this work, a methodology for easy design and real Implementation of microprocessors is proposed, in order to provide students with a user-friendly tool. Simple designs of microprocessors are exposed to the students at the beginning, rising the complexity gradually toward a final design with two processors integrated in an FPGA; each of which has an independent memory system, and are intercommunicated with a unidirectional serial channel.

     2   MULTIPROCESSOR
       Multiprocessor system consists of two or more
Connect processors that are capable of communicating. This can be done on a single chip where the processors are connected typically by either a bus. Alternatively, the multiprocessor system can be in more than one chip, typically connected by some type of bus, and each chip can then be a multiprocessor system. A third option is a multiprocessor system working with more than one computer connected by a network, in which each
Computer can contain more than one chip, and each chip can contain more than one processor.
     A parallel system is presented with more than one task, known as threads. It is important to spread the workload over the entire processor, keeping the difference in idle time as low as possible. To do this, it is important to coordinate the work and workload between the processors. Here, it is especially crucial to consider whether or not some processors are special-purpose IP cores. To keep a system with N processors effective, it has to work with N or more threads so that each processor constantly has something to do. Furthermore, it is necessary for the processors to be able to communicate with each other, usually via a shared memory, where values that other processors can use are stored. This introduces the new problem of thread safety. When thread safety is violated, two processors (working threads) access the same value at the same time. Some methods for restricting access to shared resources are necessary. These methods are known as thread safety or synchronization. Moreover, it is necessary for each processor to have some private memory, where the processor does not have to think about thread safety to speed up the processor. As an example, each processor needs to have a private stack. The benefits of having a multiprocessor are as follows:
1. Faster calculations are made possible.
2. A more responsive system is created.
3. Different processors can be utilized for different Tasks. In the future, we expect thread and process parallelism to become widespread for two reasons: the nature of the Applications and the nature of the operating system. Researchers have therefore proposed two alternatives Micro architectures that exploit multiple threads of Control: simultaneous multithreading (SMT) and chip multiprocessors (CMP). Chip multiprocessors (CMPs) use relatively simple.
    Single-thread processor cores that exploit only moderate amounts of parallelism within any one thread, while executing multiple threads in parallel across multiple processor cores. Wide-issue superscalar processors exploit instruction level parallelism (ILP) by executing multiple instructions from a single program in a single cycle. Multiprocessors (MP) exploit thread-level parallelism (TLP) by executing different threads in parallel on Different processors.

3 EMBEDDED SYSTEMS AND SYSTEM DESIGN       
3.1 Characteristics of Embedded Systems
The earliest embedded systems were banking and transaction processing systems running on mainframes and arrays of disks. The design of such a system entails hardware- software co-design: given the expected number and type of transactions to be made in a day, a hardware configuration must be chosen that will support the expected traffic and a software design must be created to efficiently make use of that hardware. When microprocessors are used to create specialized, low-cost products, engineering costs must be reduced to a level commensurate with the cost of the underlying hardware.
    Because multiprocessors can be used in such a wide range of products, embedded systems may need to meet widely divergent criteria. Examples of embedded systems include:
•   simple appliances, such as microwave ovens, where the multiprocessor provides a friendly interface and advanced features;
•   an appliance for a computationally intensive task, such as laser printing;

Read More: Click here...