Hendrik Moens

VNF-P: A model for efficient placement of virtualized network functions

Network Functions Virtualization (NFV) is an upcoming paradigm where network functionality is virtualized and split up into multiple building blocks that can be chained together to provide the required functionality. This approach increases network flexibility and scalability as these building blocks can be allocated and reallocated at runtime depending on demand. The success of this approach depends on the existence and performance of algorithms that determine where, and how these building blocks are instantiated. In this paper, we present and evaluate a formal model for resource allocation of virtualized network functions within NFV environments, a problem we refer to as Virtual Network Function Placement (VNF-P). We focus on a hybrid scenario where part of the services may be provided by dedicated physical hardware, and where part of the services are provided using virtualized service instances. We evaluate the VNF-P model using a small service provider scenario and two types of service chains, and evaluate its execution speed. We find that the algorithms finish in 16 seconds or less for a small service provider scenario, making it feasible to react quickly to changing demand.

Design and evaluation of a hierarchical application placement algorithm in large scale clouds

As the requirements and scale of cloud environments increase, scalable management of the cloud is needed. Centralized solutions lack scalability and fully distributed management systems only have a limited overview of the system. One of the often-studied problems in cloud environments is the application placement problem, used to decide where application instances are instantiated and how many resources to allocate to the instances. In this paper a general approach is introduced for using centralized cloud resource management algorithms in a hierarchical context, increasing the scalability of the management system while maintaining a high placement quality. The management system itself is executed on the cloud, further increasing scalability and robustness. The proposed method uses aggregation techniques to generate input values for a centralized application placement algorithm which is run in all management nodes. Decoupling ensures management nodes can function independently. Subsequently, we compare the performance of hierarchical application placement method with that of a fully centralized algorithm. The results show that a solution, within 5% of the optimum placement when using the centralized algorithm, can be achieved hierarchically in less than 25% of the time needed for execution of the centralized algorithm.

Hierarchical network-aware placement of service oriented applications in Clouds

In cloud environments, resources can be requested on-demand when they are needed. A cloud management system is responsible for determining which physical machines are responsible for processing the requests. The problem of determining which servers are used for which services is referred to as the Cloud Application Placement Problem (CAPP), and multiple criteria such as cost and number of migrations must be taken into account. When applications are constructed as a collection of communicating services, such as in Service-Oriented Architectures, it becomes important to take the underlying network properties into account when these placement decisions are made. In this paper, we propose an Integer Linear Programming (ILP) formulation for the CAPP, which optimizes multiple criteria such as cost, latency and number of migrations between subsequent invocations by using multiple optimization criteria. We also present hierarchical algorithms based on particle swarm optimization and genetic algorithms to solve the CAPP. These algorithms are be executed within a management hierarchy, which reduces the amount of information needed for the algorithms to function, increasing scalability of the management system. Finally, we evaluate the hierarchical algorithms by comparing them to an optimal algorithm based on the ILP formulation.

Developing and managing customizable Software as a Service using feature model conversion

In recent years, there has been a growing interest in cloud technologies. Using current cloud solutions, it is however difficult to create customizable multi-tenant applications, especially if the application must support varying Quality of Service (QoS) guarantees. Software Product Line Engineering (SPLE) and feature modeling techniques are commonly used to address these issues in non-cloud applications, but these techniques cannot be ported directly to a cloud context, as the common approaches are geared towards customization of on-premise deployed applications, and do not support multi-tenancy. In this paper, we propose an architecture for the development and management of customizable Software as a Service (SaaS) applications, built using SPLE techniques. In our approach, each application is a composition of services, where individual services correspond to specific application functionalities, referred to as features. A feature-based methodology is described to abstract and convert the application information required at different stages of the application life-cycle: development, customization and deployment. We specifically focus on how development feature models can be adapted ensuring a one-to-one correspondence between features and services exists, ensuring the composition of services yields an application containing the corresponding features. These runtime features can then be managed using feature placement techniques. The proposed approach enables developers to define significantly less features, while limiting the amount of automatically generated features in the application runtime stage. Conversion times between models are shown to be in the order of milliseconds, while execution times of management algorithms are shown to improve by 5 to 17% depending on the application case.

Feature placement algorithms for high-variability applications in cloud environments

While the use of cloud computing is on the rise, many obstacles to its adoption remain. One of the weaknesses of current cloud offerings is the difficulty of developing highly customizable applications while retaining the increased scalability and lower cost offered by the multi-tenant nature of cloud applications. In this paper we describe a Software Product Line Engineering (SPLE) approach to the modelling and deployment of customizable Software as a Service (SaaS) applications. Afterwards we define a formal feature placement problem to manage these applications, and compare several heuristic approaches to solve the problem. The scalability and performance of the algorithms is investigated in detail. Our experiments show that the heuristics scale and perform well for systems with a reasonable load.

Algorithms for advance bandwidth reservation in media production networks

Media production networks connect different actors such as production houses, broadcasters and advertisers over a large geographical area. The substrate network in a media production process can be shared among many users simultaneously. These networks often have predictable traffic, meaning that the timing and bandwidth requirements of data transfers are generally known at least hours or even days in advance. Using this knowledge, an efficient advance reservation mechanism can be deployed to improve the performance in terms of resource utilization and cost. In such mechanisms, to improve the performance, defining strategies to deal with unforeseen failures in the network is required. In this paper, we propose a resilient advance bandwidth reservation algorithm for media production networks. To enable a quick response to sudden changes such as failures in the network, we rely on a protection mechanism, which means that backup paths are found in advance, before any failure happens in the network. The proposed scheme aims at minimizing the resource usage of backups, while it guarantees 100% recovery against any single link failure. We have performed an availability analysis of the proposed scheme considering an optical transport network and compared it with the non-resilient approach. The results show that the proposed scheme performs reasonably well and enhances the availability significantly compared to the non-resilient approach.

Feature-based application development and management of multi-tenant applications in clouds

In recent years, there has been a rising interest in cloud computing, which is often used to offer Software as a Service (SaaS) over the Internet. SaaS applications can be offered to clients at a lower cost as they are usually multi-tenant: many end users make use of a single application instance, even when they are from different organisations. It is difficult to offer highly customizable SaaS applications that are still multi-tenant, which is why these SaaS applications are often offered in a one size fits all approach. In some application domains applications must be highly customizable, making it more difficult to migrate them to a cloud environment, and losing the benefits of multi-tenancy. In this paper we compare multiple approaches for the development and management of highly customizable multitenant SaaS applications, and present a methodology for developing and managing these applications. We compare two approaches, an application-based approach focusing on deploying multiple multi-tenant applications variants, and a feature-based approach where applications are composed out of multi-tenant services using a service oriented architecture. In addition, we also discuss a hybrid approach combining properties of both. We conclude that the feature-based approach results in the fewest application instances at runtime resulting in more multi-tenancy.

A scalable approach for structuring large-scale hierarchical cloud management systems

In recent years, the scale of clouds and networks has increased greatly. It is important to ensure that the management systems used in these environments can scale as well. A centralized system does not scale well, while for distributed approaches, it is difficult to maintain an overview of the global system state. In hierarchical management systems, nodes at a low level in the hierarchy have a detailed view of a small part of the network, while higher-level nodes have a less detailed view of larger parts of the network. This makes hierarchical management systems well suited for large scale systems. The structure of such a hierarchical system should however be impacted by the management system for which it is used, as various properties such as the number of child nodes, tree depth and the distance between nodes can impact the performance of the management system. In this paper, we describe the Scalable Hierarchical Management Framework (SHMF), a scalable approach for constructing a hierarchical management system, suitable for large-scale cloud environments, that automatically optimizes its structure in function of its overlying management system. We evaluate the approach based on the requirements for the cloud application placement problem.

Characterizing the performance of tenant data management in multi-tenant cloud authorization systems

Multi-tenancy leads to improved efficiency, improved scalability, and lower costs. With the recent evolution of Cloud Computing and Software-as-a-Service (SaaS) in particular, a flexible and scalable multi-tenant architecture is becoming highly important. In multi-tenant applications, each tenant has its own users and administrators and tenants even tend to be divided into multiple subtenants. As the number of tenants grows, the number of users and amount of data grows, thus a scalable architecture for the access control system is needed. The question arises how to distribute the users and data over multiple database instances. In this paper we present a hierarchical data management approach, taking performance metrics into account, for structuring the storage of tenant data in large multi-tenant environments. We introduce a logical representation of the tenants, the tenant tree, and make a mapping to the physical storage by introducing three models for load-balancing. Next, we focus on how to efficiently locate the required data and introduce multiple search approaches. We characterize the impact on the performance both theoretically and experimentally. Experiments confirm that the theoretical analysis is in line with the experimental results. When the amount of data increases significantly, dividing the data over multiple datastores in an efficient way will eliminate the overhead and lead to a performance gain, especially if most of the data is located at the leaf nodes of the tenant tree.

Scalable user data management in multi-tenant cloud environments

The rise of cloud computing and its elastic, on-demand resource provisioning introduces the need for a flexible and scalable multi-tenant architecture. In a multi-tenant application every tenant (client) makes use of shared application instances, but each tenant typically has its own user data. The shared application instance behaves like a private instance by guaranteeing both data separation and performance separation for every tenant. As the number of tenants increases, the amount of data grows. A scalable solution for the storage is needed, allowing tenant data to be divided over multiple database instances, but taking into account performance isolation and custom data assurance policies. In this paper we introduce an abstraction layer for achieving high scalability for the storage of tenant data. This layer uses data allocation algorithms to determine an acceptable allocation of tenant data to different databases. We describe a mathematical model for the allocation of tenant data which can be optimized using existing linear programming techniques, and introduce the BDAA-n and FDAA, two algorithms that will find an optimal allocation of data by iterating over the possible permutations. The proposed solutions are evaluated based on their flexibility, complexity and efficiency. The flexibility of the BDAA and FDAA makes them easy to customize and extend to fit most scenarios, but the algorithms will achieve best results for tenants with a limited number of subtenants. Linear programming is an alternative for tenants with a higher number of subtenants, but the customizability of the algorithm for specific use cases is limited due to the need for linear functions.