Session 3.2 - Enterprise Software on the Cloud

Chair: Mor Sagmon, SAP Research Israel

Session Abstract

How do you take a robust, mission critical business application, and deliver it from the cloud? New models of software development, delivery and consumption pose new challenges for software providers and businesses alike. How will storage and virtual machines be managed, migrated and realized? In this session we'll glimpse into some of these challenges and new ideas stemming out of the research labs of SAP and IBM, and enjoy a sneak preview under the hood of SAP's new in-memory DB: HANA.



Biography: Mor Sagmon Business Development, SAP Research, Israel, SAP

Mor Joined SAP in 2000 and held various positions in engineering, consulting, customer relationships, business development in the public sector and SAP Research. Mor is overlooking public-funded research projects in EMEA and leads the SAP University Alliances Program in Israel. Prior to SAP, Mor held various positions in the private industrial sector, including the position of a CIO in a manufacturing company. In the past, Mor engaged in systems analysis, training, book co-authoring and development. Mor holds a B.A. in Computer Science from the Technion, the Israeli prestige technology education and research institution.


SAP HANA – In-memory Database Architecture

Speaker: Chaim Bendelac, Vice President, SAP Labs Israel.

Lecture Abstract: In-memory computing is changing the way businesses are run. An accelerating demand for real-time analytics on increasingly bigger structured and unstructured data sets on one hand, and the introduction of multi-core CPUs and large main memories on the other, have established an inflection point in Enterprise Database Design in recent years. We will describe and illustrate these technology trends by presenting the architecture of SAP HANA, SAP's In-memory Enterprise Database.


Biography: Chaim Bendelac Chief Engineer with the SAP HANA team at SAP, researches system performance and optimization. Prior to SAP, Chaim managed several teams as Engineering Director and Director of Product Management at Cisco Systems.  Chaim was founder and VP Engineering of InfoGear Technologies, a startup focused on web appliances that was acquired by Cisco and in charge of a compiler development  at National Semiconductor. Chaim has a M.Sc. in Computer Science from Stony Brook University  and an executive MBA from the Kellogg/Recanati business schools.  

Lecture Title: The Lego Cloud: Using Virtualization to Disaggregate the Datacenter

Speaker: Dr. Eliezer Levy, SAP Research Israel.

Lecture Abstract: The advances in fast interconnect and virtualization technologies open a range of options when building systems in general and IaaS and PaaS cloud platforms in particular. In principle, fast remote memory access and transparent compute virtualization enable departing from sharing a physical host by several virtual machines (VMs) into building flexible multi-host VMs from modular building blocks of compute and memory resources.  These features can be provided transparently by extending the open and standard Linux kernel and KVM hypervisor without requiring any change from applications. As first result of this research we demonstrate the optimization of live migration of VMs by transferring and accessing memory pages transparently regardless of their physical location. Then we gradually introduce the scenarios of memory aggregation across the network, instantaneous flash cloning of VMs, and finally Distributed Shared Memory (DSM). All these scenarios make use of the basic capabilities that we have developed in Linux/KVM.  The outcome is savings in TCO and increased operational flexibility, while providing extreme compute and memory machines; all this without requiring changes from applications and with minimal performance degradation.


Biography: Eliezer Levy established and manages the SAP Research group in Israel. Prior to that, he was the Chief Architect of the Small Businesses business unit in SAP. Eliezer has over 20 years of experience in various research, development and management roles in SAP, Intel, Tandem Computers and an Israeli startup. Eliezer has a Ph.D. in Computer Sciences from the University of Texas at Austin, and a B.Sc. in Computer Science from the Technion. Among his active areas of research are databases, systems and enterprise computing.  

Lecture Title: Advancements in Cloud-Based Storage

Speaker: Dr. Dalit Naor, IBM Haifa Research Lab.

Lecture Abstract: The explosion of personal and organizational digital data, and of fixed content in particular, is one of the key characteristics of our decade. Cloud Storage, which refers to the pooling of multiple geographically dispersed storage resources into what appears to be a single storage resource, provides an infrastructure that can potentially address many of these challenges. But will enterprises adopt cloud-based storage solutions to host their data?

In this talk I will discuss what it takes to develop an enterprise-worthy yet cost-effective cloud storage infrastructure. I will present the VISION Cloud European funded project. Led by IBM Haifa, the VISION Cloud project is building a powerful infrastructure for reliable and effective delivery of data-intensive storage services, facilitating the convergence of IT, media and telecommunications. In the talk we will present the main concepts and innovations that are the foundations of the VISION Cloud architecture, and focus on the object store service that is developed by IBM. I will demonstrate the capabilities of the object service by examples taken from different domains such as Telco, Media and Healthcare.


Biography: Dr. Dalit Naor manages the Storage Research group at IBM Haifa Research Lab
( Dalit holds M.Sc and PhD degrees from the University of California, Davis, and a B.Sc. degree from the Technion - Israel Institute of Technology, all in computer science. She is specializing in developing advanced features for next generation storage systems, functions and interfaces, with focus on Cloud Storage, long term digital preservation, data reduction techniques, object storage and power-efficient storage systems. Dalit worked on security-related technologies, including internet security, content protection, and storage security, and her past academic research includes bioinformatics and combinatorial optimization.