Turning down the lamp: Software specialisation for the cloud

Compiling is not only running a program that interprets a given source file collection. In the compilation process, it is important the immediacy and way of display the results, and the user interaction. In this research it is proposed the creation of a unified platform (.Cloud) that supports editing, compiling and running applications in multiple languages and that can execute directly into the user's browser without installing any plugins, making it independent of the platform and operating system used. .Cloud will be independent of the platform on which it runs, which will favor mainly to devices with limited resources, both hardware and platform software. The use of a Cloud tool of this type also facilitates the work group within computing projects, allowing multiple programmers working on the same data with optimized workflow.

… on Cluster Computing …, 2009

Cloud computing systems fundamentally provide access to large pools of data and computational resources through a variety of interfaces similar in spirit to existing grid and HPC resource management and programming systems. These types of systems offer a new programming target for scalable application developers and have gained popularity over the past few years. However, most cloud computing systems in operation today are proprietary, rely upon infrastructure that is invisible to the research community, or are not explicitly designed to be instrumented and modified by systems researchers.

Proceedings of the 2017 Symposium on Cloud Computing

Distributed computing remains inaccessible to a large number of users, in spite of many open source platforms and extensive commercial offerings. While distributed computation frameworks have moved beyond a simple map-reduce model, many users are still left to struggle with complex cluster management and configuration tools, even for running simple embarrassingly parallel jobs. We argue that stateless functions represent a viable platform for these users, eliminating cluster management overhead, fulfilling the promise of elasticity. Furthermore, using our prototype implementation, Py-Wren, we show that this model is general enough to implement a number of distributed computing models, such as BSP, efficiently. Extrapolating from recent trends in network bandwidth and the advent of disaggregated storage, we suggest that stateless functions are a natural fit for data processing in future computing environments. CCS CONCEPTS • Computer systems organization → Cloud computing; • Computing methodologies → Distributed programming languages;

2014

Abstract—Cloud Platforms allow programmers to write ap-plications that run in the cloud, or use services from the Cloud, or both while abstracting the essence of scalability and distributed processing. With the emergence of Clouds as a nascent architecture, we need abstractions that support emerging programming models. In recent years, Cloud Computing has led to the design and developement of diverse programming models for Massive Data Processing and Compute Intensive Applications. We survey different programming models for the Cloud, including the popular MapReduce Model and others which improve upon its shortcomings. Further, we look at models which are promising alternatives to these parallel programming models.

2014

Cloud computing has not resulted in a fundamental change to the underlying operating systems. Rather, distributed applications are built over middleware that provides high-level abstractions to exploit the cloud’s scale and elasticity. This middleware conjoins many general purpose OS instances. Others have demonstrated that a new operating system built specifically for the cloud can achieve increased efficiency, scale and functionality [11, 14]. However, this work does not take into account the way applications are being deployed in cloud environments. In particular, entire physical or virtual machines are being dedicated to run a single application, rather than concurrently supporting many users and multiple applications. In this paper we introduce a new model for distributed applications that embraces a reduced role of the OS in the cloud. It allows for the construction of application-driven compositions of OS functionality wherein each application can employ its own customized op...

2015

All Operating Systems provide interface to users and inherent applications running on a system. Today with the advancement in the field of cloud computing, its services are in high demand. The Cloud is becoming an active paradigm of computation through services. However, such services do not prove to be very efficient in slow networks. Thus, the paper intends to achieve faster response of services provided by the Cloud infrastructure through the use of a new mechanism of servicing client requests. The focus of the paper is to realize most operating system functions at the client side over the Cloud infrastructure using the virtualization technique. It also proposes to overcome the issues present in the current infrastructure service and resulting in a faster service.

2020 43rd International Convention on Information, Communication and Electronic Technology (MIPRO), 2020

Increased complexity of network-based software solutions and the ever-rising number of concurrent users forced a shift of the IT industry to cloud computing. Conventional network software systems commonly based on monolithic application stack running on costly physical single-purpose servers are affected by significant problems of resource management, computing power distribution, and scalability. Such implementation is restricting applications to be reduced to smaller, independent services that can be more easily deployed, managed, and scaled dynamically; therefore, embellishing environmental uniformity across development, testing, and production. Current cloud-based infrastructure frequently runs on containers placed in Kubernetes or Docker-based cluster, and the system configuration is considerably different compared to the environment prevailed with common virtualizations. This paper discusses the usage of GraalVM, a polyglot high-performance virtual machine for JVM-based and other languages, combined with new Kubernetes native Java tailored stacked framework named Quarkus, formed from enhanced Java libraries. Moreover, our research explores GraalVM's creation of native images using Ahead-Of-Time (AOT) compilation and Quarkus' deployment to Kubernetes. Furthermore, we examined the architectures of given systems, various performance variables, and differing memory usage cases within our academic testing environment and presented the comparison results of selected performance measures with other traditional and contemporary solutions.

2008

Although they allow easy and cost-effective use of a wide range of machines, the programming interface and behavior of general-purpose Operating Systems (OS) often fail to meet, or even conflict with, the specific desires of High-Performance Computing (HPC) applications, such as low preemption or control over memory and I/O management. That often leads to poor performance. On the other hand, hypervisors are more and more commonly used on top of those OSes for various reasons, such as ease of dedicated environment deployment or load balancing. In contrast to the usual unix process model, hypervisors provide their guests with kernel-level facilities. In this paper, we show how an HPC application and its execution environment can be embedded within a lightweight guest domain, alongside a domain that runs a conventional OS which is only used for administrative purpose. That permits the execution environment to take advantage of kernel-level facilities to improve performance, which would...

2013

Executed during years 2010-2013, Cloud Software is a Finnish research program, whose goal has been to significantly improve the competitive position of Finnish software intensive industry in global markets in the field of cloud computing. More than 30 Finnish IT companies and research organizations have participated in the program, and the output of the program has been considerable both in terms of scientific publications as well as industry transformation towards using cloud computing techniques. This book contains selected contributions by the Cloud Software Program partners focusing on innovative algorithms, applications, and tools to develop new services and applications to be deployed in public and private cloud infrastructures. These are needed to fully utilize the concept computing as a utility as promised by cloud computing. In addition to the actual contributions, the first chapter of the book contains an introduction to cloud computing from the point of view of software development. For the rest of the book, each chapter tackles a concrete problem relevant to software development for cloud infrastructures that originated in the program company partners. We want to thank all the participants in the Cloud Software program for their e↵orts, shared experiences, and insights during these four years that have been materialized in the contents of the book. Specially, we want to thank Dr. Janne Järvinen, focus area director, Professor Veikko Seppänen, academic coordinator, Dr. Tua Huomo, program coordinator, and Dr. Pauli Kuosmanen, Digile CTO, for leading the program. In addition, we wish to thank Hans Ahnlund and Mika Laurila (ECE Ltd.), Marc Englund (Vaadin Ltd.), Sami Helin (Steeri Oy), Jarno Kallio (Packet Video Finland), Risto Laurikainen (CSC), and Heikki Nousiainen (F-Secure) for performing reviews for the di↵erent chapters that constitute the book. Finally, we also wish to thank all the contributors of this book for their dedicated e↵orts both in the program itself as well as in reporting the results.