|
|
Newsletter December 2019 - Issue 5
|
|
Message from the Coordinator
Hello everybody,
Welcome to the last issue of the PrEstoCloud Newsletter. PrEstoCloud is nearing completion. Including the preparation phase for this project, we jointly developed and shaped our ideas and innovations on the topic of "Cloud Computing" for almost four years. During this period, we developed an architecture, that manages cloud and fog resources proactively, enabling to reach the extreme edge of the network. Also, three pilot projects in different domains demonstrates the power of PrEstoCloud's new approach. In addition, we have produced an impressive number of publications and appearances at conferences, giving new impetus to the academic world. Looking back today, we can see that the selection of partners, but especially the structuring of our project of nine work packages, including the three use cases, was chosen purposefully. Numerous innovations were initiated and point to a promising exploitation of the project results. I wish us all every success in this respect.
Facing the end of the official project phase I would like to take the opportunity to thank all consortium partners for their inspiring and effective cooperation. In numerous face-to-face meetings and working groups, all have contributed their respective technological experiences and have further developed them into a successful dialogue between science and industry. I would also like to thank especially the Project Officer Luis C. Busquets Pérez, who facilitated the administrative handling of our project and all involved members and responsibles of the European Commission in supporting us during the whole project phase.
I wish you happy holidays and a wonderful 2020. Enjoy reading!
With best regards,
Dirk Mayer, Software AG Project Coordinator
|
|
|
PrEstoCloud Use Cases
PrEstoCloud mainly focussed on three industry domains that
have been covered by PrEstoCloud pilots. They demonstrate
the power of PrEstoCloud's new approach:
Logistics Pilot
The CVS logistics application is capable of tracking
their vehicles movements, the geographic locations where
driving dynamics (e.g. sudden acceleration, hard
braking, aggressive right turn and aggressive left turn)
happened and what types of driving dynamics occurred
during the trips. The CVS logistics centre is now able
to profile their drivers’ behaviour and associated
statistics. Moreover, the CVS logistics application
implemented by the PrEstoCloud project knows if any
driver is dangerously manoeuvring at the moment on
situations when many driving dynamics are currently
happening.
Through the CVS logistics application implemented in the
PrEstoCloud solution, stakeholders such as operators of
the logistics centre, vehicles owners, drivers and
insurance companies can track, maintain, optimise, and
manage the telematics data and logistical assets such as
their fleet of vehicles.
CVS takes advantage of the following benefits:
-
Increasing fleet safety which is an objective
for the CVS Mobile. For example, telematics data
is useful for evaluating and hence correcting
driver behaviour that insurance companies
hopefully begin to offer discounts for the CVS
Mobile.
-
Telematics data allows the CVS Mobile to
optimise maintenance schedules since developing
custom maintenance schedules for each vehicle is
necessary to identify when and where performance
issues arise in a fleet vehicle. For example,
the CVS Mobile can be able to determine when a
fleet vehicle may need an oil change, new
braking pads or a filter swap, and hence the CVS
fleets will be capable of paying for maintenance
more efficiently.
-
Minimising infrastructure costs through
technologies by which services and data can be
efficiently provisioned, monitored and migrated
across federated cloud, fog and edge resources.
Surveillance Pilot
The surveillance systems, that capture video and audio
in enterprise facilities and public places, produce
massive amounts of data while operating in a 24/7 mode.
There is an increasing need to process, on the fly, such
huge video and audio data streams to enable a quick
summary of “interesting†events that are happening
during a specified time frame at a particular location.
The smart surveillance application, developed by
ADITESS, is a combination of audio/video analytics with
visualization detecting security events and informing
the appropriate security personnel. A novel and adaptive
architecture builds on top of a distributed computing
paradigm and is ideal for smart surveillance systems
that can utilize resources at cloud, fog and edge. The
surveillance use cases demonstrates how PrEstoCloud can
ensure the availability of services, system integrity
and real-time detection of illicit activity despite the
increased workload, unexpected failure of devices or
computational components.
Through the PrEstoCloud Project, we enabled a novel and
adaptive architecture that built on top of a distributed
computing paradigm which is ideal for smart surveillance
systems that uses resources at cloud, fog and edge.
OPPORTUNITIES / CHALLENGES
-
Video analysis at the edge of the network is
part of
new concepts for cloud computing and increases
company
profile.
-
Cost savings by deploying efficient processing
at the edge especially for video.
-
Cost efficiency- use of legacy Hardware (HW)
-
Unmanned Aerial Vehicle (UAV) integration under
the same framework
CAPABILITIES
-
More efficient data analysis at the edge
providing a distributing surveillance system
-
Only analytics outputs and some functions run on
cloud
-
Better utilization of resources at the edge of
the network
-
Higher security features
-
Less storage in the cloud
-
Less data transfer to the cloud
-
Higher trust and privacy having less processing
in the cloud
Mobile Journalism Pilot
LiveU and N.Amram Technologies have developed a product for
Mobile Journalism
users. Nowadays, when life is very dynamic, anyone can
become an eyewitness of some remarkable event.
At the same time, a high level of e-communication between
people via communication tools encourage everyone to become
a newsmaker. This trend creates new opportunities
for news companies in optimization their operation by using
Mobile Journalism. The application developed by LiveU and
NAM allows news companies, professionals and volunteers
to work together on news making: to analyse live video
streams by using video analytics, communicate in private
virtual rooms, exchange video feeds. With PrEstoCloud
platform,
all these services are optimised in terms of hardware
resources and costs which allows news companies to minimize
their expenses and at the same time keep the high quality of
news feeds.
|
|
|
PrEstoCloud Architecture and Technologies
Fog Context Analytics Component
Fog computing environments are unstable and diverse. Edge
devices with different hardware and software configuration
have different processing capabilities and performance.
Factors such as the available RAM, the type of memory chips,
the number of CPU cores, frequency, type and generation, the
current battery level (when the device lowers CPU
clock frequency in order to lower energy consumption) or the
speed of storage devices (HD, SSD, Flash) may affect the QoS
(Quality of Service, e.g. total execution time)
of application and services in different ways. Considering
the aforementioned characteristics of fog computing, with
the term context we refer to any information
(such as CPU, memory or utilization, network type and
traffic, battery state, software and hardware configuration,
etc.) that can help to infer conclusions about current
and future state of fog devices.
A clear and accurate classification of context is helpful to
uncover, understand, manipulate and sort out a variety of
sensor-generated data to uncover useful information about
the status of a computing resource and help DevOps make
infrastructure management decisions. Many machine learning
methods and mathematical representations of fog
infrastructures
have been used to convert low-level sensor data into
higher-level context, which can facilitate decision- making.
Indicative methods used for context classification include
Hidden
Markov Models, Markov Chains, Bayesian Networks, Nearest
Neighbor, Time Series, Threshold based Learning and Gaussian
Mixture Models.
ICCS has analysed key methods of context classification and
prediction and the application of these methods to
facilitate fog infrastructure management. In the context of
PrEstoCloud,
ICCS has developed a Fog Context Analytics component, which
can support infrastructure operators in case where there is
an increased demand for management automation, robustness
and overall management complexity reduction to achieve good
quality of service and serve demands of different devices
and applications.
The software component is available at GitLab.
Mobile Offload Processing
Mobile based real-time processing network based on
cloud-edge communication layer was developed and
deployed by JSI for general PrEstoCloud platform and
supporting variety of situations including all three use
cases of the project. Video is available
here.
|
|
|
PrEstoCloud on Tour
PrEstoCloud Webinars
On November 29, and September 4 the PrEstoCloud Webinars
took place. Participants were informed about PrEstoCloud
project in general, the microservices architecture, proposed
key influencing auto-scaling factors, and the empirical
evaluation. Of course, there was an opportunity for
interaction and discussion.
International Workshop on Autonomic High Performance
Computing (AHPC 2019)
The PrEstoCloud project team work benefited from the
discussion within the 2019 International Workshop on
Autonomic High Performance Computing (AHPC 2019) organized
and chaired by Prof. Françoise Baude (CNRS/I3S). The
workshop AHPC 2019 was co-located with the 2019
International Conference on High Performance Computing and
Simulation (HPCS 2019), at the Grand Hotel Malahide in
Dublin, Ireland, on June 15-19, 2019.
The workshop focused on autonomic computing with an emphasis
on self-manageable systems that exhibit self-configuration,
self-healing, self-protection, self-optimization, and/or
self-monitoring, which is ubiquitous in the world of
distributed applications, such as embedded/IoT devices,
cloud, and big data applications, such as targeted by the
PrEstoCloud project.
Telematics Conference CEEurope in Prague
Telematics Conference CEEurope in Prague hosted
PrEstoCloud project the second time. On 26th September,
Demo of current CVS edge to cloud technology platform
solution impressed many participants including
commercial telematics ones. It showed a new wave and
opportunities for further development within fleet
management operations, driver’s profiling not to mention
connecting professionals with approach to upgraded
insurance concept possibilities.
PrEstoCloud at SIGKDD Conference in Anchorage
SIGKDD project showcase track on August 4-8 in Anchorage,
Alaska brought together a diverse community of researchers
in machine learning and data analytics with the aim of
promoting innovation and stare-of-the-art research in
presented areas. JSI and CVS mobile introduced poster and
orally presented accepted paper on Efficient Real-Time Big
Data Processing at the Edge of the Network.
Upcoming Events
Hannover Messe 2020
PrEstocloud belongs to one of the research projects being
shown at Hannover Messe 2020 at the booth of Software AG.
From 20 - 24 April 2020, Software AG's research manager will
be there to share the research findings to visitors from the
manufacturing industries and the energy sector.
|
|
|
Dissemination Material
In the final phase of the project, many publications emerged
from the teamwork. Here you will find a list of new
dissemination activities:
Scientific papers from the last project year:
Scientific papers from the last project year
- Key influencing factors of the Kubernetes auto-scaler for computing-intensive microservice-native cloud-based applications. Advances in Engineering Software 140, Elsevier, 2019. Get the paper!
Taherizadeh, Salman; Grobelnik, Marko
- SWITCH workbench: A novel approach for the development and deployment of time-critical microservice-based cloud-native applications. Future generation computer systems 99, Elsevier, 2019. Get the paper!
Stefanic, Polona; Cigale, Matej; Jones, Andrew C.; Knight, Louise; Taylor, Ian; Istrate, Cristiana; Suciu, George; Ulisses, Alexandre; Stankovski, Vlado; Taherizadeh, Salman; et al.
- Dynamic multi-level auto-scaling rules for containerized applications. The computer journal 62(2), Oxford University Press, 2019. Get the paper!
Taherizadeh, Salman; Stankovski, Vlado
- Measuring the Amazon Web Services (AWS) WAN Infrastructure at CoRes 2019 (Rencontres Francophones sur la Conception de Protocoles, l'Evaluation de Performance, et l'Expérimentation des Réseaux de Communication), 2019. Get the paper!
Jacquemart, Quentin; Vitali, Alessandro; Urvoy-Keller, Guillaume
- nferring the Deployment of Top Domains over Public Clouds using DNS Data at the 3rd Network Traffic Measurement and Analysis Conference (TMA 2019). Get the paper!
Jacquemart, Quentin; Pigout, Clément; Urvoy-Keller, Guillaume
- Efficient Real-Time Big Data Processing at the Edge of the Network – The PrEstoCloud project. KDD 2019 Project Showcase Track (KDD 2019), ACM, Alaska, United States, 2019. Get the paper!
Taherizadeh, Salman; Novak, Blaz; Vagaja, Sebastjan; Komatar, Marija; Grobelnik, Marko
Videos
A clear overview of the project and the use cases can be
found in the
videos.
Blog Posts
On the PrEstoCloud
Blog you will find some topics and
questions that research partners have dealt with in the last
three years.
|
|
|
Meet our Project Team
The multinational research consortium consists of very
experienced researchers and technology developers, who have
already provided parts of the system that will be used as a
basis for developing the PrEstoCloud platform. The
consortium unites a total of eleven partner organizations
from industry and research. Software AG has been entrusted
with the leadership of the consortium
|
|
|
|
Join us on Facebook, Twitter and Linkedin
We invite you to use our online networking tools to make
connections,
exchange scientific ideas, and keep informed about the
latest research
findings. On Facebook, Twitter and Linkedin you will enter
rooms for
exchange of experiences, among professionals and
stakeholders - a great
opportunity to connect with a greater scientific an
industrial
community.
|
|
|
|