Runtime management to explore alternative implementations

Loading...

来自 Politecnico di Milano 的课程

FPGA computing systems: Background knowledge and introductory materials

8 个评分

Politecnico di Milano

FPGA computing systems: Background knowledge and introductory materials

8 个评分

This course is for anyone passionate in learning how a hardware component can be adapted at runtime to better respond to users/environment needs. This adaptation can be provided by the designers, or it can be an embedded characteristic of the system itself. These runtime adaptable systems will be implemented by using FPGA technologies. Within this course we are going to provide a basic understanding on how the FPGAs are working and of the rationale behind the choice of them to implement a desired system. This course aims to teach everyone the basics of FPGA-based reconfigurable computing systems. We cover the basics of how to decide whether or not to use an FPGA and, if this technology will be proven to be the right choice, how to program it. This is an introductory course meant to guide you through the FPGA world to make you more conscious on the reasons why you may be willing to work with them and in trying to provide you the sense of the work you have to do to be able to gain the advantages you are looking for by using these technologies. We rely on some extra readings to provide more information on the topic covered in this course. Please NOTE that most of the time, these documents are provided through the IEEE Xplore Digital Library, which means that, to access them, you have to have a valid IEEE subscriptions, either does by yourself or through your university/company. The course has no prerequisites and avoids all but the simplest mathematics and it presents technical topics by using analogizes to help also a student without a technical background to get at least a basic understanding on how an FPGA works. One of the main objectives of this course is to try to democratize the understanding and the access to FPGAs technologies. FPGAs are a terrific example of a powerful technologies that can be used in different domains. Being able to bring this technologies to domain experts and showing them how they can improve their research because of FPGAs, can be seen as the ultimate objective of this course. Once a student completes this course, they will be ready to take more advanced FPGA courses.

从本节课中

Towards Partial Dynamic Reconfiguration and Complex FPGA-based systems

The reconfiguration capabilities of FPGAs give the designers extended flexibility in terms of hardware maintainability. FPGAs can change the hardware functionalities mapped on them by taking the application offline, downloading a new configuration on the FPGA (and possibly new software for the processor, if any) and rebooting the system. Reconfiguration in this case is a process independent of the execution of the application. A different approach is the one that considers reconfiguration of the FPGA as part of the application itself, giving it the capability of adapting the hardware configured on the chip resources according to the needs of a particular situation during the execution time. In this case we are referring to this reconfiguration as dynamic reconfiguration and the reconfiguration process is seen as part of the application execution, and not as a stage prior to it. This module illustrates a particular technique, which is extending the previous two, that has been viable for most recent FPGA devices, Partial Dynamic Reconfiguration. To fully understand what this technique is, the concepts of reconfigurable computing, static and dynamic reconfiguration, and the taxonomy of dynamic reconfiguration itself must be analyzed. In this way partial dynamic reconfiguration can be correctly placed in the set of system development techniques that it is possible to implement on a modern FPGA chip.

How to deal with the Reconfiguration runtime overhead3:42

Recurring modules to reuse them to reduce the Reconfiguration time3:50

Partial Reconfiguration to reduce the Reconfiguration runtime overhead5:38

Runtime management to explore alternative implementations5:54

Bitstreams relocation6:46

Bitstreams relocation and virtual homogeneity3:02

与讲师见面

Marco Domenico Santambrogio
Associate Professor
DEIB - Dept. of Electronics, Information and Bioengineering

Object code implemented as hardware components do not always guarantee the best performance.

We know that we need a runtime: this can be realised by means of a standalone solution

which is providing an optimized solution for the specific problem,

but it requires large implementation efforts, since it is necessary to rewrite the whole application

each time the underlying hardware changes.

Or it can be implemented via an operating system which is, on the other hand,

oriented towards increasing the level of abstraction and flexibility of the system,

and simplifying module requests by providing a set of high-level system calls.

Within this context, the development of an operating system for reconfigurable devices

makes the whole system more flexible, and increases the level of abstractions

for the final user and this is why in the following example I’m going to assume

we are going to use a GNU/Linux based runtime.

Now, let us supposed to have a scenario which involves two algorithms for cryptography:

the Advanced Encryption Standard (AES) and the Data Encryption Standard (DES)

The AES algorithm is a cryptography algorithm adopted as a United States Government standard,

and it works on 128-bit data blocks using different key lengths.

In the considered implementation,

only the encryption operation using a 128-bit key is considered.

On the other hand, the DES algorithm is the predecessor of the AES

and employs a shorter key, which is 56-bit long, to encrypt or decrypt a 64-bit block of input data.

Both a hardware and a software implementation are available for each cryptography algorithm.

The system is implemented on a System-on-Chip, where the software implementations are executed

on the processor implemented on the FPGA logic and executing at 100 MHz.

Furthermore, they are handled by the OS as high-priority processes,

and therefore their execution times are not affected by most of the tasks

which are always active in the system.

The hardware implementations of the AES and the DES functionalities are implemented

as reconfigurable functional unites, and they can be placed on the FPGA on demand.

The two cryptographic functionalities are registered via a centralized manager,

which makes them available to userspace applications.

Software applications may request one of the two algorithms, and they are required

to specify the size of the input text to be encrypted.

A large text must be split into fixed-size blocks which are encrypted separately,

and thus each block of data requires a separate execution of the algorithm.

When an AES or a DES request is received by the centralized manager,

the operating system evaluates whether the hardware or the software implementation should be chosen.

The selection of the implementation is based on the input size

and on the number of available slots to allocate the IP-Core

These figures show the execution of the hardware and software implementations of AES and DES.

As expected, the latency of the software executions increases linearly with the input size,

with only small variations due to the time sharing policy that is imposed by the Operating System

on the processor.

The hardware implementation exploits the inherited parallelism of hardware circuits,

and hence performs much better than the software counterpart.

However, the execution of the IP-Core version of the algorithm requires the dynamic reconfiguration

of the hardware module itself, which introduces a nearly constant delay.

Given the large size of the IP-Cores - whose partial bitstreams reaches the 500 kiloBytes -,

the delay can be estimated in the order to 150 milliseconds,

which makes the hardware implementation inefficient for input data of small size.

The IP-Core outperforms the software implementation when the input size exceeds a threshold,

which depends on the particular algorithm.

As an example, results show that a software execution of the AES algorithm is more efficient

if less than 40 blocks of data are processed.

On the other hand, the reconfiguration and the execution of the DES IP-Core

does not provide a significant improvement if less than 80 blocks are encrypted.

This information can be employed by the OS - in addition to area availability -

in order to select the proper implementation at runtime.

Furthermore, the table shows how the throughput of the software and the hardware implementations

of the AES and the DES algorithms is influenced by the input data size,

assuming that the IP-Core must be configured on the device.

The throughput of the two software versions is nearly constant, whereas it linearly increases

for the hardware implementations, which explains the poor performances

of dynamic reconfiguration for small input data.

Module caching can vastly improve the performances, since it removes the reconfiguration overhead.

The idea behind modules caching is quite simple:

caching a module means that once it has been configured on the FPGA

we are not going to remove it to reuse it in the future.

A caching mechanism can be successfully applied in this case study when a requested functionality

is already configured in an FPGA slot and it is currently unused.

If the corresponding IP-Core is immediately available, the throughput of the AES algorithm

is constant and it is equal to 436 kBytes/s, while the DES algorithm

shows a constant throughput of 246 kBytes/s.

This is just an example.

We can have several applications that may vary performance based on several factors

and the size of the inputs is just one of them.

This is exactly the reason why we need to have, not only an hardware architecture capable

to be adapted at runtime but also a runtime that may monitor the system performance

and take decision based on them.

探索我们的目录

免费加入并获得个性化推荐、更新和优惠。

Coursera 致力于普及全世界最好的教育，它与全球一流大学和机构合作提供在线课程。

© 2018 Coursera Inc. 保留所有权利。

通过 App Store 下载

通过 Google Play 获取