Fiber-based architecture for NFV cloud databases

Abstract

The telco industry is gradually shifting from using monolithic software packages deployed on custom hardware to using modular virtualized software functions deployed on cloudified data centers using commodity hardware. This transformation is referred to as Network Function Virtualization (NFV). The scalability of the databases (DBs) underlying the virtual network functions is the cornerstone for reaping the benefits from the NFV transformation. This paper presents an industrial experience of applying sharednothing techniques in order to achieve the scalability of a DB in an NFV setup. The special combination of requirements in NFV DBs are not easily met with conventional execution models. Therefore, we designed a special shared-nothing architecture that is based on cooperative multi-tasking using user-level threads (fibers). We further show that the fiber-based approach outperforms the approach built using conventional multi-threading and meets the variable deployment needs of the NFV transformation. Furthermore, fibers yield a simpler-to-maintain software and enable controlling a trade-off between long-duration computations and real-time requests.

In Broker We Trust: A Double-Auction Approach for Resource Allocation in NFV Markets

“…NFV changes the traditional rules for how network operators manage their infrastructure as software instances separate from the hardware platform by using the proven virtualization technology [11]. For example, one example of an open platform for NFV (OPNFV), the Huawei E9000 blade server, is widely applied in industry and is intended to facilitate the commercial adoption of NFV applications [12].…”

AIDA-DB: A Data Management Architecture for the Edge and Cloud Continuum

“…Another example is the network functions implemented through virtual means, i.e., Virtual Network Functions (VNFs). To meet demands, VNFs can be easily scaled, which leads to stateful functions such as Network Address Translation (NAT), load balancing, firewalls, among others, to require transactional guarantees to manage their shared state, as data sharding is not always possible [16]. State management performance here is especially important, as network functions must display considerably low delays [2].…”

Analyzing the impact of system architecture on the scalability of OLTP engines for high-contention workloads

“…For instance, Caldera [2] uses message passing to build OLTP engines on non-cache-coherent hardware. Thread migration is being used to build virtualization-friendly OLTP engines [12]. Orthrus [36] combines message passing with functional partitioning and transaction pre-execution to improve scalability under high-contention workloads.…”