Search results for: data-centers

Authors: Nihed Bahria El Asghar, Imen Jouili, Mounir Frikha

Abstract:

Inter-datacenter transport network is very bandwidth and delay demanding. The data transferred over such a network is also highly QoS-exigent mostly because a huge volume of data should be transported transparently with regard to the application user. To avoid the data transfer failure, a backup path should be reserved. No re-routing delay should be observed. A dedicated 1+1 protection is however not applicable in inter-datacenter transport network because of the huge spare capacity. In this context, we propose a survivable virtual network with minimal backup based on network coding (1 ⊕ 1) and solve it using a modified Dijkstra-based heuristic.

Keywords: network coding, dedicated protection, spare capacity, inter-datacenters transport network

Procedia PDF Downloads 419

Authors: Wanyi Zhu, Xia Ming, Huafeng Wang, Junda Chen, Lu Liu, Jiangwei Jiang, Guohua Liu

Abstract:

Anomaly detection is a first and imperative step needed to respond to unexpected problems and to assure high performance and security in large data center management. This paper presents an online anomaly detection system through an innovative approach of ensemble machine learning and adaptive differentiation algorithms, and applies them to performance data collected from a continuous monitoring system for multi-tier web applications running in Alibaba data centers. We evaluate the effectiveness and efficiency of this algorithm with production traffic data and compare with the traditional anomaly detection approaches such as a static threshold and other deviation-based detection techniques. The experiment results show that our algorithm correctly identifies the unexpected performance variances of any running application, with an acceptable false positive rate. This proposed approach has already been deployed in real-time production environments to enhance the efficiency and stability in daily data center operations.

Keywords: Alibaba data centers, anomaly detection, big data computation, dynamic ensemble learning

Procedia PDF Downloads 170

Authors: Jörg Lenhardt, Wolfram Schiffmann, Jörg Keller

Abstract:

While the feature sizes of recent Complementary Metal Oxid Semiconductor (CMOS) devices decrease the influence of static power prevails their energy consumption. Thus, power savings that benefit from Dynamic Frequency and Voltage Scaling (DVFS) are diminishing and temporal shutdown of cores or other microchip components become more worthwhile. A consequence of powering off unused parts of a chip is that the relative difference between idle and fully loaded power consumption is increased. That means, future chips and whole server systems gain more power saving potential through power-aware load balancing, whereas in former times this power saving approach had only limited effect, and thus, was not widely adopted. While powering off complete servers was used to save energy, it will be superfluous in many cases when cores can be powered down. An important advantage that comes with that is a largely reduced time to respond to increased computational demand. We include the above developments in a server power model and quantify the advantage. Our conclusion is that strategies from datacenters when to power off server systems might be used in the future on core level, while load balancing mechanisms previously used at core level might be used in the future at server level.

Keywords: power efficiency, static power consumption, dynamic power consumption, CMOS

Procedia PDF Downloads 198