FRuIT: The Federated RaspberryPi Micro-Infrastructure Testbed
Lead Research Organisation:
University of Glasgow
Department Name: School of Computing Science
Abstract
The introduction of ubiquitous low-cost, low-power compute devices, e.g. in the Internet of Things (IoT) is fundamentally changing the computational landscape. Although we already see the benefits provided by special purpose IoT devices, the true capability is only realised when we can re-purpose large numbers of distributed devices as part of a much larger federated service. This is captured in the FRUIT project hypothesis:
"Massive aggregation of low-cost, low-power, commodity infrastructure can form an efficient and
effective compute fabric for key distributed applications."
The participating institutions and external collaborators will build a distributed UK-wide federated testbed of lightweight compute nodes capable of exploring the research issues that arise from such a dynamic infrastructure.
This is in contrast to the existing centralised datacentre approach which occupies acres of land, cost millions of dollars; consumes megawatts of power and relies on a huge global network bandwidth. These trends are not sustainable and it is the emergence of low cost hardware such as the Raspberry Pi that will let us explore better solutions. We envisage a testbed comprising tens of thousands of nodes, geographically distributed and capable of challenging remote off-grid scenarios. In order to facilitate this we introduce the concept of a micro-datacentre, a physically small, low power and low cost compute cluster that can be pushed to the edge of distributed networks, fundamentally changing the current model of the centralised datacentre. We have already built prototype micro-data centres at multiple sites , for example a mini-HPC cluster (Southampton University), a scale model cloud datacentre (Glasgow and Liverpool John Moores University ),
and a decentralized sensing and communication platform for ultra-remote internet blackspots (Cambridge University)
This project is about connecting together our isolated micro-datacenters to produce a federated, geo-distributed testbed. We currently have hundreds of individual nodes and intend to grow this to tens of thousands across the UK academic network. The key challenge in this expansion will be the management of the underlying infrastructure, particularly given complexities in the networking fabric. For example intermittent connectivity, firewalls, and low bandwidth connections. We will use lightweight management and orchestration software, combined with software-defined networking infrastructure to manage our federated testbed.
Through this work, we aim to demonstrate how resource-constrained micro-data centres can be harnessed as generic platforms to run virtualised, network-wide services in a resource-efficient and still high-performing manner, and enable the 'as-a-Service' paradigm over low-cost low-power federated infrastructures.
"Massive aggregation of low-cost, low-power, commodity infrastructure can form an efficient and
effective compute fabric for key distributed applications."
The participating institutions and external collaborators will build a distributed UK-wide federated testbed of lightweight compute nodes capable of exploring the research issues that arise from such a dynamic infrastructure.
This is in contrast to the existing centralised datacentre approach which occupies acres of land, cost millions of dollars; consumes megawatts of power and relies on a huge global network bandwidth. These trends are not sustainable and it is the emergence of low cost hardware such as the Raspberry Pi that will let us explore better solutions. We envisage a testbed comprising tens of thousands of nodes, geographically distributed and capable of challenging remote off-grid scenarios. In order to facilitate this we introduce the concept of a micro-datacentre, a physically small, low power and low cost compute cluster that can be pushed to the edge of distributed networks, fundamentally changing the current model of the centralised datacentre. We have already built prototype micro-data centres at multiple sites , for example a mini-HPC cluster (Southampton University), a scale model cloud datacentre (Glasgow and Liverpool John Moores University ),
and a decentralized sensing and communication platform for ultra-remote internet blackspots (Cambridge University)
This project is about connecting together our isolated micro-datacenters to produce a federated, geo-distributed testbed. We currently have hundreds of individual nodes and intend to grow this to tens of thousands across the UK academic network. The key challenge in this expansion will be the management of the underlying infrastructure, particularly given complexities in the networking fabric. For example intermittent connectivity, firewalls, and low bandwidth connections. We will use lightweight management and orchestration software, combined with software-defined networking infrastructure to manage our federated testbed.
Through this work, we aim to demonstrate how resource-constrained micro-data centres can be harnessed as generic platforms to run virtualised, network-wide services in a resource-efficient and still high-performing manner, and enable the 'as-a-Service' paradigm over low-cost low-power federated infrastructures.
Planned Impact
We will engage in academic dissemination via research publications at targeted conference venues like NSDI, SOSP, VEE, Infocom, IEEE Cloud, also appropriate journals like Transactions on Networks, Computer Networks.
e.g. Cox and Johnson are editing a special issue of the Electronics journal on `Raspberry Pi Technology'.
We will organize community workshops, separately for both academic and general enthusiast audiences.
We have significant experience in this area, e.g. Glasgow academics have presented invited seminars and hands-on demos at 11 workshop events over the past 3 years. These include BCS events, Pi Jam events, PhD summer schools etc.
Further, many academics, industrialists and enthusiasts across the world consult us for guidance when building their own micro-datacenters. They follow our progress on the Glasgow project blog at http://raspberrypicloud.wordpress.com
We will undertake public engagement of the massive online Raspberry Pi community across the world. We will release open source instructions for all our infrastructure, hardware blueprints, tool configuration and software source code. Initially the aim will be for others to federate with us directly, later we will encourage them to clone their own distributed systems.
We intend to achieve long-term impact via standardization of configuration and deployment protocols for micro-datacenters. We expect to engage with industry and standards bodies to convert our work into official standards. e.g. Colin Perkins interacts regularly with the IETF standardization process.
In terms of the testbed sustainability, we intend to grow over the funded project to a number of sites across JANET - at least 10. We have secured agreement from Heriot Watt and UCL to build Raspberry Pi testbeds there already.
We recognise that new generations of low-power embedded devices are coming onstream all the time. What will the Raspberry Pi 3 look like? We appreciate the need to be able to retarget our software stack to these other devices too. However once the standards are in place and our demonstrator testbed is live, we expect that the open source community will become involved in porting our work to other infrastructures.
e.g. Cox and Johnson are editing a special issue of the Electronics journal on `Raspberry Pi Technology'.
We will organize community workshops, separately for both academic and general enthusiast audiences.
We have significant experience in this area, e.g. Glasgow academics have presented invited seminars and hands-on demos at 11 workshop events over the past 3 years. These include BCS events, Pi Jam events, PhD summer schools etc.
Further, many academics, industrialists and enthusiasts across the world consult us for guidance when building their own micro-datacenters. They follow our progress on the Glasgow project blog at http://raspberrypicloud.wordpress.com
We will undertake public engagement of the massive online Raspberry Pi community across the world. We will release open source instructions for all our infrastructure, hardware blueprints, tool configuration and software source code. Initially the aim will be for others to federate with us directly, later we will encourage them to clone their own distributed systems.
We intend to achieve long-term impact via standardization of configuration and deployment protocols for micro-datacenters. We expect to engage with industry and standards bodies to convert our work into official standards. e.g. Colin Perkins interacts regularly with the IETF standardization process.
In terms of the testbed sustainability, we intend to grow over the funded project to a number of sites across JANET - at least 10. We have secured agreement from Heriot Watt and UCL to build Raspberry Pi testbeds there already.
We recognise that new generations of low-power embedded devices are coming onstream all the time. What will the Raspberry Pi 3 look like? We appreciate the need to be able to retarget our software stack to these other devices too. However once the standards are in place and our demonstrator testbed is live, we expect that the open source community will become involved in porting our work to other infrastructures.
Organisations
- University of Glasgow (Lead Research Organisation)
- Illuminum Greenhouses (Collaboration)
- University of Tokyo (Project Partner)
- Raspberry Pi Foundation (Project Partner)
- Madanyu (Project Partner)
- Microsoft Research (United Kingdom) (Project Partner)
- ARM (United Kingdom) (Project Partner)
- Heriot-Watt University (Project Partner)
Publications
Zhou Q
(2020)
A Prediction-Based Model for Consistent Adaptive Routing in Back-Bone Networks at Extreme Situations
in Electronics
Zhang X
(2021)
A survey on stateful data plane in software defined networks
in Computer Networks
Nunns G
(2019)
Autonomous Flying WiFi Access Point
Dalibard V
(2017)
BOAT
Johnston SJ
(2019)
City Scale Particulate Matter Monitoring Using LoRaWAN Based Air Quality IoT Devices.
in Sensors (Basel, Switzerland)
Varghese B
(2019)
Cloud Futurology
in Computer
Johnston S
(2018)
Commodity single board computer clusters and their applications
Description | It is feasible to develop single-board compute clusters using commodity hardware like Raspberry Pi. These low-power, low-footprint, low-cost clusters have sensible use cases and deployment scenarios, including in disaster zones, developing economies, community computing, and educational learning environments. |
Exploitation Route | Our open source hardware and software can be used by others to construct their own low-power federated clusters. Several international organizations have inspected our setup and are currently engaged in replicating it. |
Sectors | Creative Economy Digital/Communication/Information Technologies (including Software) Education Electronics |
URL | https://fruit-testbed.org |
Description | A range of organizations are using our Raspberry Pi micro-cluster concept, for various end user application areas including environmental monitoring, edge computing, sensor systems. Some of the material has been incorporated into the new undergraduate level Operating Systems textbook from Arm Education - "Operating Systems Foundations with Linux on the Raspberry Pi" - which is being adopted by a number of UK universities as their core course text. |
First Year Of Impact | 2019 |
Sector | Agriculture, Food and Drink,Digital/Communication/Information Technologies (including Software),Education,Culture, Heritage, Museums and Collections |
Impact Types | Cultural Societal Economic |
Description | Models for Edge and In-Network Computation |
Amount | £400,000 (GBP) |
Organisation | Rakuten Institute of Technology |
Sector | Academic/University |
Country | Japan |
Start | 08/2020 |
End | 08/2024 |
Description | RAEng Seed Funding (GCRF) (Eiko Yoneki at Cambridge) |
Amount | £20,000 (GBP) |
Organisation | Royal Academy of Engineering |
Sector | Charity/Non Profit |
Country | United Kingdom |
Start | 02/2018 |
End | 02/2019 |
Title | LoRaWAN for smart city IoT deployments |
Description | Data set to accompany paper titled LoRaWAN for smart city IoT deployments: A long term evaluation |
Type Of Material | Database/Collection of data |
Year Produced | 2020 |
Provided To Others? | Yes |
Impact | none as yet |
URL | https://zenodo.org/record/3572514 |
Title | PyonAir: An open design, open source, air quality monitor for community driven particulate matter sensing. |
Description | This dataset present some of the data recorded by two low-cost PM sensors, a Plantower PMS5003 and a Sensirion SPS030 located at Southampton AURN reference station on the 2nd December 2019 between 17:00 and 22:00 during a fire that occurred in the city. It also present the data from the Fidas 200, averaged every 15min also located at the AURN station. The file SPS_PMS_fire.csv contain the following rows: date pm25 - PM 2.5 mass concentration in ug/m 3 sensor - Serial number of the sensor site - name of the location of the sensor The file fidas_15min.csv contains the following rows: date PM2.5 - PM 2.5 mass concentration in ug/m 3 |
Type Of Material | Database/Collection of data |
Year Produced | 2020 |
Provided To Others? | Yes |
URL | https://zenodo.org/record/3691933 |
Title | PyonAir: An open design, open source, air quality monitor for community driven particulate matter sensing. |
Description | This dataset present some of the data recorded by two low-cost PM sensors, a Plantower PMS5003 and a Sensirion SPS030 located at Southampton AURN reference station on the 2nd December 2019 between 17:00 and 22:00 during a fire that occurred in the city. It also present the data from the Fidas 200, averaged every 15min also located at the AURN station. The file SPS_PMS_fire.csv contain the following rows: date pm25 - PM 2.5 mass concentration in ug/m 3 sensor - Serial number of the sensor site - name of the location of the sensor The file fidas_15min.csv contains the following rows: date PM2.5 - PM 2.5 mass concentration in ug/m 3 |
Type Of Material | Database/Collection of data |
Year Produced | 2020 |
Provided To Others? | Yes |
URL | https://zenodo.org/record/3691932 |
Title | Testing Smart City environmental monitoring technology using small scale temporary cities |
Description | This is the data used for: S. J. Johnston et al., "Testing Smart City environmental monitoring technology using small scale temporary cities," 2019 IEEE 5th World Forum on Internet of Things (WF-IoT), 2019, pp. 578-583, doi: 10.1109/WF-IoT.2019.8767274. Abstract: Exposure to particulate matter has been identified as a major health problem worldwide. Established measurement techniques require equipment costing many thousands of dollars and specialist expertise to maintain. Ongoing research is investigating the use of low cost <$300 sensors to enable greater temporal-spatial density of readings to be taken. There are questions about the suitability and reliability of these low-cost sensors, queries which can be addressed by deploying and evaluating the sensors in a real world application. We propose festival site as small scale cities to enable a short term deployments and evaluation of sensors. We present data from these devices and experiences gained from using a festival site as a substitute for a city. Files: 1 - timeLapse: mp4 file presenting a time lapse of the measurements realised during the festival 2 - sensor_data: csv file containing the 5 min averaged data from all the sensors deployed and their coordinates used to generate the graph and the maps in the paper 3 - workshop.pdf: instructions to run the workshop and code used for the workshop |
Type Of Material | Database/Collection of data |
Year Produced | 2019 |
Provided To Others? | Yes |
URL | https://zenodo.org/record/2267379 |
Title | flexible sensor evaluation board. |
Description | Open source hardware description files to build a PCB for the Raspberry Pi to create a flexible sensor evaluation board. |
Type Of Material | Computer model/algorithm |
Year Produced | 2019 |
Provided To Others? | Yes |
Impact | Potential collaborations with EU partner researchers. |
URL | https://zenodo.org/record/2545263 |
Description | ESpot: Early Detection of Pest and Disease in Greenhouses in Africa |
Organisation | Illuminum Greenhouses |
Country | Kenya |
Sector | Private |
PI Contribution | Deployment of a smart greenhouse monitoring solution using raspberry pi network connected to the cloud. |
Collaborator Contribution | Visiting various green houses in Kenya and examined the efficiency of the deployed devices and data collection. |
Impact | A prototype of image processing system. |
Start Year | 2019 |
Title | PiStack PCB Design |
Description | The files specify a printed circuit board (PCB) design that enables Raspberry Pi boards to be inter-connected in a straightforward manner, allowing them to be integrated into micro-scaled compute clusters. |
Type Of Technology | Physical Model/Kit |
Year Produced | 2018 |
Impact | The PCBs manufactured from this specification enable the easy construction of stacked clusters of Raspberry Pi nodes, facilitating single-board computer clusters. There is a growing research community in this area, which is attracting commercial interest. Our boards are useful infrastructure building blocks for such communities. |
URL | https://eprints.soton.ac.uk/416906/ |
Title | Plantower Particulate Sensor Python interface |
Description | A basic python interface for interacting with the plantower Particulate Monitoring sensors. This code has been tested with the following devices: Plantower PMS5003 Plantower PMS7003 Plantower PMSA003 |
Type Of Technology | Software |
Year Produced | 2018 |
Open Source License? | Yes |
Impact | Widespread use online, democratizing particulate sensor monitoring in urban areas. |
URL | https://pypi.org/project/plantower/ |
Title | System software for Federated Raspberry Pi Micro-Infrastructure Testbed |
Description | A software configuration, deployment and management infrastructure for hosted Linux instances running on single-board compute clusters, like the FRuIT Raspberry Pi federated cluster. |
Type Of Technology | Software |
Year Produced | 2018 |
Open Source License? | Yes |
Impact | The development of the FRuIT infrastructure, as a conceptual piece of hardware |
URL | https://github.com/fruit-testbed/ |
Description | An Afternoon on IoT for Smart Cities |
Form Of Engagement Activity | Participation in an activity, workshop or similar |
Part Of Official Scheme? | No |
Geographic Reach | Local |
Primary Audience | Industry/Business |
Results and Impact | This afternoon event on the topic of IoT for Smart Cities included 9 presentations from academia and industry, along with 7 supporting posters to showcase research. Around 70 visitors from local industry and academia attended the event. |
Year(s) Of Engagement Activity | 2019 |
URL | https://www.c-iot.ecs.soton.ac.uk/event-06march19 |
Description | BUGALERT: Pest and Disease Monitoring in Greenhouses in Kenya, Talk on 2019.7.12, University of Cambridge Plant Science Department |
Form Of Engagement Activity | A talk or presentation |
Part Of Official Scheme? | No |
Geographic Reach | Local |
Primary Audience | Postgraduate students |
Results and Impact | Overview of interdisciplinary research project describing implementation of research for the UN Sustainable Development Goals in ODA-target country (Kenya). Led to useful conversations regarding wider applications of the research. |
Year(s) Of Engagement Activity | 2019 |
Description | Digital Epidemiology: Understanding Epidemic Spread using Human Contact Networks |
Form Of Engagement Activity | A talk or presentation |
Part Of Official Scheme? | No |
Geographic Reach | National |
Primary Audience | Professional Practitioners |
Results and Impact | A presentation at the International Centre for Theoretical Physics entitled ' Digital Epidemiology: Understanding Epidemic Spread using Human Contact Networks' - cross-disciplinary presentation about Computing Science research. |
Year(s) Of Engagement Activity | 2017 |
Description | Facilitation of Maker event |
Form Of Engagement Activity | Participation in an open day or visit at my research institution |
Part Of Official Scheme? | No |
Geographic Reach | Regional |
Primary Audience | Undergraduate students |
Results and Impact | I ran a hardware hack event with students, showing them how to program simple microcontrollers to do internet-of-things style sensing. We used BBC micro:bit devices, programming in Python |
Year(s) Of Engagement Activity | 2017 |
URL | https://www.gla.ac.uk/events/makerspace/ |
Description | Glasgow Science Festival Participation (Explorathon) |
Form Of Engagement Activity | Participation in an activity, workshop or similar |
Part Of Official Scheme? | No |
Geographic Reach | Regional |
Primary Audience | Public/other audiences |
Results and Impact | A robot racing game, demonstrating autonomous vehicle technology to the general public. |
Year(s) Of Engagement Activity | 2017 |
URL | http://www.explorathon.co.uk/ |
Description | Networkshop47 Talk |
Form Of Engagement Activity | A talk or presentation |
Part Of Official Scheme? | No |
Geographic Reach | National |
Primary Audience | Professional Practitioners |
Results and Impact | A presentation about Cyber Security to university IT managers and technical staff, in collaboration with JISC. Led to increased awareness of low-level cyber security issues with IoT style devices. |
Year(s) Of Engagement Activity | 2019 |
URL | https://www.jisc.ac.uk/events/networkshop47-09-apr-2019/programme |
Description | Picycle scheme for Raspberry Pi recycling |
Form Of Engagement Activity | Engagement focused website, blog or social media channel |
Part Of Official Scheme? | No |
Geographic Reach | National |
Primary Audience | Public/other audiences |
Results and Impact | We are encouraging people to donate their old single board computers (e.g. Raspberry Pi devices) to our FRuIT research project, by sending them to us at a FREEPOST address. We then use these machines to build our cluster. |
Year(s) Of Engagement Activity | 2017,2018 |
URL | https://picycle.org |
Description | Presentation @ Networkshop46 entitled FRµIT: Raspberry Pi Clusters and other adventures in networking research |
Form Of Engagement Activity | A formal working group, expert panel or dialogue |
Part Of Official Scheme? | No |
Geographic Reach | National |
Primary Audience | Professional Practitioners |
Results and Impact | an overview of our FRuIT project, highlighting its technical novelties and relevance for JISC infrastructure. |
Year(s) Of Engagement Activity | 2018 |
URL | https://www.jisc.ac.uk/networkshop |
Description | Strathclyde Raspberry Pi Day |
Form Of Engagement Activity | A talk or presentation |
Part Of Official Scheme? | No |
Geographic Reach | Regional |
Primary Audience | Public/other audiences |
Results and Impact | Outreach presentation about Raspberry Pi based research at Glasgow University. |
Year(s) Of Engagement Activity | 2017 |
URL | https://www.strath.ac.uk/science/physics/bringingphysicstoyou/raspberrypiday/ |
Description | Talk at the Arm Research Summit 2017 |
Form Of Engagement Activity | Participation in an activity, workshop or similar |
Part Of Official Scheme? | No |
Geographic Reach | International |
Primary Audience | Professional Practitioners |
Results and Impact | Outreach style talk about our research project, delivered to employees at Arm and other international researchers. |
Year(s) Of Engagement Activity | 2017 |
URL | https://developer.arm.com/research/summit/previous-summits/2017/speakers |
Description | presentation at Workshop on Heterogeneous and Low-Power Data Center technologies (HeLP-DC) |
Form Of Engagement Activity | A talk or presentation |
Part Of Official Scheme? | No |
Geographic Reach | International |
Primary Audience | Industry/Business |
Results and Impact | Description of a custom peer-to-peer software updating framework we have developed as part of our FRuIT project, which may be applicable to a wider set of systems. |
Year(s) Of Engagement Activity | 2018 |
URL | https://www.hipeac.net/events/activities/7522/help-dc/#fndtn-main |