Foraging for Better Deployment of Replicated Service Components

Abstract

Abstract. Our work focuses on distributed software services and their requirements in terms of system performance and dependability. We target the problem of finding optimal deployment mappings involving multiple services, i.e. mapping service components in the software architecture to the underlying platforms for best possible execution. We capture important non-functional requirements of distributed services, regarding performance and dependability. These models are then used to construct appropriate cost functions that will guide our heuristic optimization method to provide better deployment mappings for service components. This paper mainly focuses on dependability. In particular, a logic enabling replication management and deployment for increased dependability is presented. To demonstrate the feasibility of our approach, we model a scenario with 15 services each with different redundancy levels deployed over a 10-node network. We show by simulation how the deployment logic proposed is capable to satisfy replica deployment requirements.

Laying Pheromone Trails for Balanced and Dependable Component Mappings

“…We observed that the smaller domains, e.g. d 3 , d 4 , were overloaded compared to the rest due to φ 1 , but generally replicas were placed quite evenly, showing that cooperation between the species worked.…”

“…However, Fernandez-Baca [3] showed that the general module allocation problem is NP-complete except for certain communication configurations, thus heuristics are required to obtain solutions efficiently. This paper extends our previous work to find optimal deployment mappings [4], [5] based on a heuristic optimization method called the Cross-Entropy Ant System (CEAS). The strengths of the CEAS method is its capability to account for multiple parameters during the search for optimal deployment mappings [6].…”

“…Finally, to build a cost function that satisfies our requirements with regard to Φ, while maintaining load-balancing, we formulate F () using a combination of terms, as shown in (4).…”

Ant system for service deployment in private and public clouds

“…Our previous work [8] has focused on finding efficient mappings within relatively small scale clusters; and we experimented with different pheromone encodings for improving scalability in [7]. We have targeted a decentralized solution to avoid the burden of maintaining centralized databases and to eliminate performance and dependability bottlenecks.…”

“…This value is essentially an encoding of the VM instance mapping mn,r at node n in iteration r. Hence, the pheromone database must be able to store pheromone values that encode the various deployment configurations for various services. Three possible pheromone encoding techniques are discussed and evaluated in [8]; herein the best encoding is used.…”