Boris Parák
CESNET
Network
Latest external collaboration on country level. Dive into details by clicking on the dots.
Publication
Featured researches published by Boris Parák.
Proceedings of International Symposium on Grids and Clouds (ISGC) 2014 — PoS(ISGC2014) | 2014
Boris Parák; Feldhaus Florian; Kasprzak Piotr; Srba Maik; Zdeněk Šustr
OCCI (Open Cloud Computing Interface) is an open protocol for management of tasks in the cloud environment focused on integration, portability and interoperability with a high degree of extensibility. It is designed to bridge differences between various cloud platforms and provide common ground for users and developers alike. The rOCCI framework, originally developed by GWDG, later adopted and now maintained by CESNET, was written to simplify implementation of the OCCI 1.1 protocol in Ruby and later provided the base for working client and server com- ponents giving OCCI support to multiple cloud platforms while ensuring interoperability with other existing implementations. The initial server-side component provided basic functionality and served as a proof of concept when it was adopted by the EGI Federated Cloud Task and was chosen to act as the designated virtual machine management interface. This led to further funding from the EGI-InSPIRE project, development of a full featured client and a new rOCCI-server suitable for production environment. It has also prompted further proliferation of the OCCI pro- tocol, spawned multiple connector/backend implementations and provided developers with valu- able feedback and opportunities to test their own implementations of the OCCI standard. This paper aims to provide basic information about the OCCI protocol, introduce its implementation in rOCCI, describe some of the core functionality provided by the rOCCI client and rOCCI-server along with their impact on interoperability in the cloud environment. It also briefly examines its use in the EGI Federated Cloud Task environment and explores the possibility of further integra- tion with other cloud platforms. All this with interoperability in mind. The paper also describes a carefully chosen subset of problems encountered whilst trying to provide interoperability with multiple cloud platforms through the use of the OCCI protocol, with real-world examples and chosen solutions
ieee acm international conference utility and cloud computing | 2014
Boris Parák; Zdenek ustr
In academia, many institutes have deployed IaaS cloud services in isolation, based on cloud management frameworks of their choice, making the cloudscape quite heterogeneous. Now, with the effort to federate cloud resources well underway, the challenge is not to make everyone switch to a common solution, but rather to allow everyone to keep their setup but still provide uniform access to essential services. This seems to be best achieved by implementing common standards. This paper discusses the different areas such standardization effort must address, ranging from the choice of the actual standardization approach (i.e., Choosing a recognized open standard or possibly a non-standard solution with a strong community backing) through authentication and virtual machine life-cycle management to monitoring or accounting (billing) services. The discussion is based on the authors own practical experience with implementing standard compliance in multiple cloud management frameworks, producing a real-world interoperability solution -- the rOCCI Framework - which is also introduced in the paper.
utility and cloud computing | 2017
Dalibor Klusáček; Boris Parák; Gabriela Podolníková; András Ürge
Public cloud providers are using the pay-per-use model when providing their resources to customers. Among other advantages, it allows the provider to react to changing demands, e.g., by modifying prices or by extending its physical capacities using the profit obtained. In this paper we deal with a completely different model. We describe a private scientific cloud where resources are provided to researchers for free. As we demonstrate, the absence of money means that the system must employ other mechanisms to guarantee reasonable performance and utilization. Especially, the problem of guaranteeing user-to-user fairness represents a major issue. Moreover, since there is no financial burden related to the use of cloud infrastructure, many resources can be wasted by long running idle virtual machines (VM) that their users no longer need. This leads to underutilization and resource fragmentation. This paper discusses these problems using real-life data from the CERIT Scientific Cloud and proposes several techniques to guarantee fair and efficient use of system resources. Furthermore, we present a prototype of a new experimental OpenNebula-compatible VM scheduler which was designed as a replacement for the default scheduler provided in OpenNebula distribution. Unlike the default scheduler, our new scheduler provides complex fair-sharing mechanisms as well as modular and easy-to-extend architecture to enable further development of advanced VM scheduling policies.
job scheduling strategies for parallel processing | 2017
Dalibor Klusáček; Boris Parák
Modern computing environments such as clouds, grids or HPC clusters are both complex and costly installations. Therefore, it has always been a major challenge to utilize them properly. Workload scheduling is a critical process in every production system with an unwanted potential to hamper overall performance if the given scheduler is not adequate or properly configured. Therefore, researchers as well as system administrators are frequently using historic workload traces to model/analyze the behavior of real systems in order to improve existing scheduling approaches. In this work we provide such real-life workload traces from the CERIT-SC system. Importantly, our traces describe a “mixed” workload consisting of both cloud VMs and grid jobs executed over a shared computing infrastructure. Provided workloads represent an interesting scheduling problem. First, these mixed workloads involving both “grid jobs” and cloud VMs increase the complexity of required (co)scheduling necessary to efficiently use the underlying physical infrastructure. Second, we also provide a detailed description of the setup of the system, its operational constraints and unresolved issues, putting the observed workloads into a broader context. Last but not least, the workloads are made freely available to the scientific community allowing for further independent research and analysis.
Proceedings of International Symposium on Grids and Clouds (ISGC) 2017 — PoS(ISGC2017) | 2017
Radim Janča; Dušan Baran; Boris Parák; Zdenek Sustr
The Open Cloud Computing Interface (OCCI) – a standard released by the Open Grid Forum – has found wide adoption. First came, quite naturally, server-side components. Later, with a variety of attractive computing resources made available over the standardized protocol, user-side submission tools or science gateways followed. These are usually tailored with a specific use case in mind, and – as such – they are generally capable of setting up heterogeneous data processing platforms and orchestrating complex workflows specific to their ngiven area of science. They put all this functionality at user’s disposal, often “at a single click”. On the other hand, small-scale users, backed by no software development teams, are usually relegated to the use of the simplest OCCI client in the command line. That is freely available and amply documented, but still, it by no means lowers the threshold of entry for potential federated cloud users, who often come from non-technical areas in the long tail of science. Therefore, the GUOCCI (GUI for OCCI) has been conceived as a rudimentary graphical user interface to OCCI-compliant cloud services. It is by design kept as simple as possible to address the needs of small-scale or one-off users who typically require little dynamism in their virtual resources, and who are perfectly happy to set up their virtual resources by hand, even one-by one, especially if they are offered a comprehensive graphical interface to do it. GUOCCI integrates not only with OCCI-compliant cloud sites, but also with the EGI Application Database and with authentication technologies used in academic federated clouds, namely with Virtual Organization Membership Services (VOMS). With that, the considerable resources available for instance in the EGI Federated Cloud are open up to all such small-scale or beginning users. This article introduces in greater depth the reasoning behind developing GUOCCI, and details the architecture of the product, making it into an example OCCI client implementation.
Proceedings of International Symposium on Grids and Clouds (ISGC) 2016 — PoS(ISGC 2016) | 2017
Zdenek Sustr; Michal Kimle; Lubomir Kosaristan; Boris Parák
Both the IaaS (Infrastructure as a Service) and PaaS (Platform as a Service) nmodels of providing cloud services rely on virtual appliances. In popular terms, they are images nof either bare operating systems, typically entailing popular Linux distributions, which can nbe further contextualized once users instantiate their own virtual resources, or operating systems with napplications pre-installed for use in the given platform, which may often consist of a nnumber of complimentary appliances. Such appliances must be offered to users of any cloud service -nthey are the basic units the users see and select from when they decide to procure resources in the ncloud. Understandably, cloud service providers are often expected to offer a variety of appliances. nEven in a simple IaaS scenario, users expect to see a range of OS distributions and flavours. With nPaaS, the variety is even greater. Obviously a range of appliances can be obtained from cloud nmarketplaces, but that only offsets rather than solves the problem since the challenges of maintaining ntheir appliances are the same for local cloud site administrators and marketplace maintainers alike. nThis, inevitably, means that cloud site or marketplace administrators must not only offer a nselection of appliances, but also manage them throughout their life cycle, keep them secured and nupdated, and eventually discontinue them when the time comes. It is not only cumbersome but also ninherently insecure to leave updates to the user instantiating the given appliance. On top of that, nthe ability to always offer fresh appliances to its users is a competitive advantage a cloud site nmay wish to exploit.nThis paper introduces a concept of automated periodic appliance updates in a nfederated cloud environment, alongside actual tools developed to perform that task. It also nsums up up-to-date experience with operating such tools in the European Grid Initiatives nFederated Cloud
Archive | 2017
Boris Parák; Zdeněk Šustr; Pkasprzak; František Dvořák; Maik Srba; Florian Feldhaus; Dušan Baran; Michal Kimle
2017-09-15nnAdded CHANGELOG.mdnCatching OCCI Argument Errors during parsingnMoved instance actions to async ActiveJobsnCleaned up old logging syntaxnAdd onetoken util that creates cloud tokens
international conference on cloud computing and services science | 2016
Boris Parák; Zdenźk źustr; Michal Kimle; Pablo Orviz Fernández; Álvaro López García; Stavros Sachtouris; Víctor Méndez Muñoz
The OCCI standard has been in use for half a decade, with multiple server-side and client-side implementations in use across the world in heterogeneous cloud environments. The real-world experience uncovered certain peculiarities or even deficiencies which had to be addressed either with workarounds, agreements between implementers, or with updates to the standard. This article sums up implementersâx80x99 experience with the standard, evaluating its maturity and discussing in detail some of the issues arising during development and use of OCCI-compliant interfaces. It shows how particular issues were tackled at different levels, and what the motivation was for some of the most recent changes introduced in the OCCI 1.2 specification.
international conference on cloud computing and services science | 2016
Zdenźk źustr; Diego Scardaci; Jiźí Sitera; Boris Parák; Víctor Méndez Muñoz
One of the benefits of OCCI stems from simplifying the life of developers aiming to integrate multiple cloud managers. It provides them with a single protocol to abstract the differences between cloud service implementations used on sites run by different providers. This comes particularly handy in federated clouds, such as the EGI Federated Cloud Platform, which bring together providers who run different cloud management platforms on their sites: most notably OpenNebula, OpenStack, or Synnefo. Thanks to the wealth of approaches and tools now available to developers of virtual resource management solutions, different paths may be chosen, ranging from a small-scale use of an existing command line client or single-user graphical interface, to libraries ready for integration with large workload management frameworks and job submission portals relied on by large science communities across Europe. From lone wolves in the long-tail of science to virtual organizations counting thousands of users, OCCI simplifies their life through standardization, unification, and simplification. Hence cloud applications based on OCCI can focus on user specifications, saving cost and reaching a robust development life-cycle. To demonstrate this, the paper shows several EGI Federated Cloud experiences, demonstrating the possible approaches and design principles.
Proceedings of International Symposium on Grids and Clouds 2015 — PoS(ISGC2015) | 2016
Michal Kimle; Boris Parák; Zdenek Sustr
The Open Cloud Computing Interface (OCCI) standard by OGF has become widely adopted in various cloud environments, such as the EGI Federated Cloud. It is currently supported by mainstream open source cloud management frameworks, e.g., OpenStack (through OCCI-OS) or OpenNebula (rOCCI) as well as others, less wide-spread ones. It is likewise supported by many workflow and submission tools used by user communities: VMDIRAC, JSAGA or SixSq. SlipStream to name but a few. OCCI is, however, found somewhat lacking in the availability of general-purpose clients supporting the standard. Only recently, its use was enabled only in the Ruby programming language through the rOCCI Framework, and command-line/scripting use was facilitated by the rOCCI-cli client. Naturally, there has been long-standing demand for OCCI support in other programming languages, primarily in Java. It has now been answered by the introduction of jOCCI â a native Java library implementing the OCCI class structure, rendering and transport specifications, currently in accordance with the OCCI v.1.1 specification. Provided by the same product team already producing the rOCCI framework, it is more than âjust another feature in the cloudscape.â Rather than a simple translation of the client part of rOCCI, it is a brand new product, a choice that has been made not only to make it a truly native Java library, but also to introduce additional, independent, client to validate the generic functionality of existing OCCI server applications. This work describes the new library and compares it to rOCCI in terms of design and interoperability when used against other server-side OCCI implementations. It also discusses the relative merits of implementing the client library as a fresh product, relatively separate from rOCCI, rather than just providing Java bindings for the client side of the rOCCI Framework. Finally, the future of rOCCI and jOCCI is briefly discussed in view of the emerging OCCI v.1.2 specification.