EcoCloud Annual Event scheduled for May 5th/6th will be cancelled this year due to the accelerated spread of the pandemic and the ban on mobility in the coming weeks. The health and safety of our community is paramount during these difficult times.

The EcoCloud executive committee is currently considering hosting a virtual event at a later date in 2020 in a format that would be productive for our affiliates with time zones spanning from China, Europe to the US West Coast. We are happy to report that EcoCloud recently hosted the ML4HPC workshop (as part of ASPLOS) on a virtual platform with 38 participants spanning these time zones. We have collected a survey from the participants about their experience which has so far been overwhelmingly positive, and will incorporate the feedback into future virtual events.

In the meantime, daily operations proceed with guidelines from the Swiss government and EPFL with regards to working from home.

We wish everyone safety and health in the coming weeks.

The EcoCloud Team

Keynote 1

New Compute Models to Power the Data Revolution

Rich Uhlig, Intel Labs

This talk will cover research topics on new computing models ranging from research in quantum computing, neuromorphic computing, compute-near memory, silicon photonics, trusted-execution environments, graph analytics, etc.

Rich Uhlig is the managing director of Intel Labs and an Intel senior fellow. Prior to this role, Rich was the director of Systems and Software Research in Intel Labs, where he led research efforts in virtualization, cloud-computing systems, software-defined networking, big-data analytics, machine learning and artificial intelligence. He joined Intel in 1996 and led the definition of multiple generations of virtualization architecture for Intel processors and platforms, known collectively as Intel Virtualization Technology (Intel® VT). Rich earned his Ph.D. in computer science and engineering from the University of Michigan.

Keynote 2

TBD

Ramanath Ramakrishnan, Eaton

Ramanath Ramakrishnan is executive vice president and chief technology officer for Eaton. Ramanath originally joined Eaton in 2005 as a director of corporate technology and later was named senior vice president of technology for the Industrial Sector, where he led significant technology and innovation programs for Eaton’s Aerospace, Hydraulics and Vehicle businesses. Before joining Eaton, he worked with GE for nine years, and held engineering positions with Wyman-Gordon Company, an aerospace components manufacturer. Ramanath is a certified professional engineer and holds a master’s degree in mechanical engineering from Ohio University. He is a certified Master Black Belt in Six Sigma and holds two patents.

Industrial Session

This year’s industrial session will feature prominent speakers from the IT industry including:

Speculative Execution Attacks as the New Memory Corruption: A Systems and Software Challenge for the Next Decade

Anil Kurmus, IBM Research Zurich

Spectre and Meltdown are the first of numerous transient attacks, which demonstrate that subtle CPU design and implementation issues can result in real world attacks, leaking sensitive information from systems. In this talk, we survey some Spectre-like attacks, and discuss approaches to understand, mitigate and prevent them. We draw a parallel with memory corruption and argue that a similar challenge to achieving memory safety awaits system and software researchers in designing efficient solutions to speculative execution attacks.

Anil Kurmus is a security researcher at the IBM Research Zurich laboratory. His interests are mainly on systems security, software security, microarchitecture security, storage systems, and applied cryptography, both in terms of offensive and defensive research. He holds a PhD degree (Dr.-Ing) from Technische Universitat Braunschweig (2014), and a Master’s degree (Diplome d’Ingenieur) from Telecom ParisTech (2009). His work, often in collaboration with excellent students visiting IBM Research, has been published in top systems security conferences.

Edge Computing at Huawei

Javier Picorel, Huawei

With the popularity of cloud services and intelligent devices skyrocketing, Information and Communications Technology (ICT) giants are challenged to cope with the massive data generation and user-request rates of a globally interconnected world. Huawei, as one of the leaders in the IT industry, has turned to edge computing, in which processing happens closer to where data is generated, rather than processing at the hyper-scale datacenters only. In this talk, I will discuss the opportunities and challenges of edge computing and its implications to the traditional centralized cloud models, and present real-world examples of employing edge computing at Huawei.

Javier Picorel is a research scientist leading the Infrastructure-as-a-Service (IaaS) team in Huawei’s Munich Research Center, which focuses on the hardware and software platform design and efficiency for Huawei Cloud’s portfolio, along with shaping and roadmapping the business layout and technology focus of Huawei’s public cloud. He is broadly interested in TCO-efficient cloud infrastructures through the vertical integration of hardware and software layers in the end of semiconductor technology scaling. He received a PhD in computer science from EPFL in 2017 and he is also the recipient of several awards in Huawei.

Optics for the Cloud: Challenges and Opportunities

Hitesh Ballani, Microsoft Research

Hitesh Ballani is a Senior Principal Researcher at Microsoft Research in Cambridge, UK. His research aims to build systems and networks for next-generation data centers. His current focus is on developing optical technologies for the cloud. Previously, he worked on a Quality-of-Service architecture for networked storage that ships in Windows Server. He graduated with a PhD in Computer Science from Cornell University in 2009 and then joined Microsoft.

Adversarial Samples: an Industry Research Perspective

Mathieu Sinn, IBM Research – Ireland

Adversarial samples are inputs to Deep Neural Networks (DNNs) that an adversary has tampered with in order to cause specific misclassifications. It is surprisingly easy to create adversarial samples and surprisingly difficult to defend DNNs against them. In this talk I will review the state-of-the-art and recent progress in better understanding adversarial samples and developing DNNs that are robust against them. I will then give an industry perspective on the potential threats that adversarial samples pose to security-critical applications of DNNs. Finally I will show how researchers and developers can experiment with adversarial attacks and defenses using our open-source library https://github.com/IBM/adversarial-robustness-toolbox.

Dr. Mathieu Sinn is a Research Staff Member and Manager of the AI & Machine Learning group at the IBM Research lab in Dublin, Ireland. He has a Master’s in Computer Science and a PhD in Mathematics from the University of Luebeck, Germany. He has worked on a large variety of fundamental and practical aspects of Machine Learning, with the most recent focus on robustness of AI against adversarial threats. Mathieu is an IBM certified Data Science Thought Leader, regular reviewer for top AI conferences and has served as external PhD committee member on various occasions.

Data Warehouses Are Dead, Long Live Data Warehousing!

Mehul Shah, Amazon

While data warehouses remain an important primitive for analysis, they no longer are the central information hub that decision making revolves around for enterprises. Modern data sets are too large and unstructured for traditional warehouses to handle, and traditional warehouses do not leverage the cloud. Yet, the need for warehousing data sets company-wide under a single locus of control still exists. So, instead, companies are setting up next-generation, communal information hubs – an evolution of enterprise data warehouses – dubbed “data lakes”. In this talk, I will discuss the opportunities and challenges in managing data lakes, and present AWS Lake Formation, a new service that makes it easy to setup, secure, and manage data lakes.

Mehul Shah is the GM for AWS Glue and AWS Lake Formation, and prior to that, he was a senior manager for Amazon Redshift. His expertise spans large-scale data management, distributed systems, and energy-efficient computing. His work has been published in top-tier conferences and journals and has won several awards including a Test of Time. Prior to Amazon, he was cofounder and CEO of Amiato, a startup that offered a real-time ETL cloud service. From 2004 to 2011, he was a principal research scientist at HP Labs. He received a PhD from UC Berkeley (2004) as well as MEng, BS in CS, and BS in physics degrees from MIT. He is also a member of the Sort Benchmark committee.