TANGO Newsletter #1

From now on, our goal is to share details about the main technical activities going on in TANGO Project. We want to share with the academic, research and scientific communities, state of the art knowledge in the field of how to better exploit and embrace heterogeneous architectures in software life-cycle, based on the progress made by our technical team in our quest to release a number of cool technologies and tools that we are releasing in what we call #TANGOtoolbox.



We are also proud to announce that TANGO Project has also teamed up with other EU R&D projects, launching the Heterogeneity Alliance, with the goal of advancing at this early-stage market of tools available for of heterogeneous architectures by sharing approaches and establishing a base reference to drive the development of such tools to improve all aspects of software life-cycle, thus facilitating taking advantage of the use of the new myriad of options of heterogeneous hardware available today for a growing number of scenarios.



Enjoy the contents of this newsletter!!!

• Building applications for heterogeneous hardware
• The First Release of #TANGOtoolbox.is here
• TANGO Project has launched an Initiative to form an Alliance to foster technologies related with heterogeneous hardware
• Meet TANGO Team in person

We will see you soon!!!

TANGO Project Team



 

Building applications for heterogeneous hardware

by David García (Atos)

When designing the architecture of the TANGO project we saw the necessity of a central component that helps to speed up the building process for different targeted heterogeneous architectures with different optimizations and deploy them if the different testbeds, especially in HPC use case, this ended up in the creation of Application Lifecycle Deployment Engine or ALDE. Let’s check step by step how will be shaped and which functionality and steps it is going to perform.

The user has the code of its application that can support different heterogeneous hardware, such as CPUs, GPUs, MCPs, FPGAS, etc. It submits this all to ALDE via its CLI or REST API. ALDE can start preparing from it different versions of the application. One optimized for running in CPU, other one optimized for running in CPU+GPU, other to run in CPU+MCP etc. The result of the application can be in different formats, depending on the possible platform where the application is going to run. Those formats are:

• Package – It is the simplest one, the different versions of the application are packaged into typical application package formats, such as deb, rpm, tar.gz, msi, etc.
• Image – ALDE builds a specific image, with operating system, necessary libraries plus the application ready to be used to but an small embedded device. In this case only one image is going to be built per device type.
• Container – That it could be a Docker or Singularity container. Although this could be also used in the container scenario. The main target here it is for the HPC environment. In HPC we are seeing a movement to containers to execute the applications since it makes the application more portable between HPC machines without the performance hassle of the VMs. In this case, if the user desires, apart from building the application with different optimizations depending on the accelerators available in the HPC environment, it could build the application with different libraries versions.

Once the application is built, the user could select the platform where to deploy it. Mainly in the HPC world, the user could select different configurations, an ideal or favourite one, and other alternatives. ALDE then will connect with the HPC workload manager and see the different free slots in the different nodes. Using this information will decide what it is the best reservation for the application to be launched, selecting the necessary containers to deploy everything.

TANGO project is developing right now ALDE. The first version will support Docker and Singularity containers and SLURM as workload manager. By this summer, a first release of the ALDE component can be found in our ALDE repository in Github.



 

The First Release of #TANGOtoolbox is here

by TANGO Project

We are proud to announce that the first release of our toolbox is here. This Alpha version release is comprised by most of the final components that will be released under the final open source #TANGOtoolbox.

Alpha version is focused on providing the necessary software to perform profiling of time and energy performance to obtain static benchmarking information on applications run on a heterogeneous hardware architecture.

In order to provide some examples of use of our tools, several scenarios and the related workflows of tools are described also here. All components can be used independently or in integration with other components to approach different situations, as are shown in these examples.

Each tool is accompanied by a readme file with the description of the component, installation guidelines for developers and users, and the relationship with other TANGO components.

Visit our website www.TANGO-project.eu to Access the full description of components, their open source license and links to download our software.

The upcoming Beta version will further integrate TANGO components to make it possible to explore trade-off on additional non-functional behaviors such as security, robustness and maintainability. Finally, in the Final version, the focus will be on enhancing programmer productivity.



 

TANGO Project has launched an Initiative to form an Alliance to foster technologies related with heterogeneous hardware

by Karim Djemame (University of Leeds) & Oliver Barreto (Atos)

Recent years saw the emergence of Cyber-Physical Systems (CPS), the Internet of Things (IoT) and the Smart Anything Everywhere Initiative which have the potential to transform the way we live and work. For example, the transformational impact in the long term of IoT is expected to increase significantly with mass adoption and tens of billions of things connected, generating multi-trillion-dollars in economic value. It will act as a key driver behind new business models taking advantage of the data collected by IoT, sophisticated application development platforms, analytics applied to things and distributed/parallel architectures.

As the range of applications continues to grow, e.g. CPS, IoT, connected smart objects, High Performance Computing (HPC), mobile computing, wearable computing etc. there is an urgent need to design more flexible software abstractions and improved system architectures to fully exploit the benefits of the heterogeneous platforms on which they operate. Heterogeneous parallel architectures have received considerable attention, as an efficient approach to run applications and deliver services, by combining different processor types in one system to improve absolute performance, minimise power consumption and/or lower cost. New platforms incorporating multi-core CPUs, many-core GPUs and a range of additional devices into a single solution are being introduced. These platforms are emerging in a wide range of environments spanning supercomputers to personal smartphones.  One of the challenges for future application performance lies not only with efficient node-level execution but with power consumption as well, which is a key focal point.

Although general complex engineering simulations come to mind when identifying families of applications benefiting most from heterogeneous parallel architectures, in the upcoming era of IoT and Big Data, there is significant interest in exploiting the capabilities offered by customized heterogeneous hardware such as FPGA, ASIP, MPSoC, heterogeneous CPU+GPU chips and heterogeneous multi-processor clusters all of which with various memory hierarchies, size and access performance properties. In fact, Big Data online with nearly instantaneous results demand massive parallelism and well devised divide-and-conquer approaches to exploit heterogeneous hardware, both client and server sides, to its fullest extent. Moreover, heterogeneous systems can not only handle workload with fewer and/or smaller servers (cost saving) but also slash the energy used to run certain applications, which provides clear benefits in addressing the growing interest in green solutions and the pressure to reduce the environmental impact of, e.g. data centers. A common theme across all scenarios is the need for low-power computing systems that are fully interconnected, self-aware, context-aware and self-optimizing within application boundaries.

Moreover, because the impact of hardware heterogeneity is rapidly increasing, innovative architectures, algorithms and specialized programming environments and tools are needed to efficiently use these new and mixed/diversified parallel architectures. For example, the transitions to multicore processors, GPU computing and Hardware as a Service (HaaS) cloud computing should be viewed as a single trend. As the market for heterogeneous architectures/multicore processors in embedded applications has begun to move into the product deployment stage, the need for software and the underlying programming methodologies is also increasing in parallel.

 

However, in this context, simply put, programming directly with this kind of heterogeneity is a nightmare. In order to rapidly develop this kind of solutions to help companies use them for the construction of richer experiences and innovative business models, we need to do something. As software engineers, we need something in the middle to help us deal with that heterogeneity and take advantage of high level abstractions as it has previously occurred in the past with traditional computing. Therefore, we need to develop innovative architectures, algorithms and even specialized programming environments and tools to efficiently deal with very different architectures to be more productive, obtain rapid learning curves and to build more general solutions.

 

The TANGO project is creating a toolbox (www.TANGO-project.eu/downloads) that provides tools that deal with this reality. However, there are others in the market and in the research arena that are also working in this direction. Our vision is to avoid creating yet more heterogeneity in the tools and approaches that will lead this market in the upcoming years. Therefore, instead of just fighting all from our own perspective, with our own tools, TANGO wanted to encourage collaboration and create something that will also create a link between research and the market and take advantage of potential synergies.

To drive this we initiated the

Heterogeneous Hardware & Software Alliance (H-HW&SW Alliance)

HeterogeneityAlliance.eu

This initiative has been undertaken by the members of the Research & Innovation Project TANGO (Transparent heterogeneous hardware Architecture deployment for eNergy Gain in Operation) funded under the EU H2020 programme. The initiative aims to join efforts of organizations interested in the development of future technologies and tools to advance and take full advantage of computing and applications using heterogeneous hardware.  Heterogeneous architectures have received considerable attention, as an efficient approach to run applications and deliver services, by combining different processor types in one system to improve absolute performance, minimise power consumption and/or lower cost. Consequently, the impact of such heterogeneity needs urgent consideration.

The alliance will focus on all phases of H-HW&SW, from design time, to enhanced execution, parallel programming and optimized runtime and consider a number of factors such as energy, performance, real-time, data locality and security. This will enable a new way of development and execution of next-generation applications.

The alliance main goal is to establish an organization in which anyone interested in the technological areas that are initially part of its scope can participate and collaborate pursuing a common objective: founding a common, open-source and an extendable set of technologies and tools around the development for heterogeneous hardware, which can be viable for mass adoption utilizing technologies created by the alliance members and attractive, easy to use and broader in scope and value. The alliance is currently working on the establishment of a reference architecture with current alliance members and is creating an online catalogue of tools and technologies. If you are interested, please contact us on www.heterogeneityalliance.eu/contact.



 

Meet TANGO Team in person

by TANGO Project

TANGO initiative is focused not only on developing great tools to deal with heterogeneity, but also to spread the word and get in touch with others involved in developing software that uses heterogeneous hardware or that create software tools for them.

To showcase our technology and tools we are always searching for events in which we can interact with this type of audience, so if you are interested in knowing the benefits that TANGO Toolbox (www.TANGO-project.eu/downloads) can provide to you, don’t miss the opportunity to meet with us in person in the events that we have scheduled our participation in the upcoming months. In these events we facilitate attendees the opportunity to talk with our technical experts so you can take a hands-on look at our tools and learn more about how to improve your workflows.

• HiPEAC Computing Systems Week: HiPEAC is organizing the upcoming Autumn Computing Systems Week in Stuttgart, Germany, on October 25th-27th. This CSW will have a focus towards innovations on automotive, mobility and related areas, with 4 main tracks: "Horizontal" Multicore session (organized by Oliver Bringmann, Tübingen, t.b.c.); HIPP; TETRAMAX + related IAs session (organized by Rainer Leupers, RWTH Aachen); and Automotive + smart mobility session (organized by Jürgen Becker, KIT). More info at https://www.hipeac.net/csw/2017/stuttgart/ 

• HiPEAC Conference 2018, Manchester: The HiPEAC conference is the premier European forum for experts in computer architecture, programming models, compilers and operating systems for embedded and general-purpose systems. The 13th HiPEAC conference will take place in Manchester, UK from Monday, January 22 to Wednesday, January 24, 2018. Associated workshops, tutorials, special sessions, several large poster session and an industrial exhibition will run in parallel with the conference. More info at https://www.hipeac.net/2018/manchester/