Hybrid Data Center Research Articles

Dynamic capacity sharing with multi-wavelength integrated transmitters in hybrid datacenter networks

In datacenters, bursty and unevenly distributed traffic may lead to serious network performance degradation. Various methods, including reconfigurable optical circuit switching (OCS), traffic control techniques, valiant load balancing (VLB), and so on, have been proposed to solve this problem. Based on these solutions, our method makes a trade-off between cost and performance. In this paper, we propose to use multi-wavelength tunable transmitters in our previously proposed modular arrayed waveguide grating (AWG)-based interconnection network. We discuss how the multiple wavelengths can be shared in the network and then propose a computational model to study its blocking probability. Closed-form equations for low network load cases are also derived to provide the analytical expression for the blocking probability. We verify the accuracy of our computational model through simulations. Comparing the blocking probability of networks with and without multi-wavelength integrated transmitters, we show that network performance can be considerably improved after replacement. When traffic burstiness is 1.25 and traffic skewness is 0.08, the blocking probability is reduced from 0.14 to 3.60×10−3 after replacing in each sending module one fixed laser with multi-wavelength tunable transmitters with four wavelengths. Furthermore, we also discuss how different factors influence the blocking probability and the maximum load with the given network performance requirement.

Application of the Multi-Threading Method and Python Script for the Automate of Network

In recent times, there has been a noticeable surge in the inclination towards the implementation of network automation solutions. This trend is driven by the objective of optimizing network availability within the context of hybrid data center networks, which are becoming increasingly prevalent. Reliability, performance, scalability, and minimal resource overhead are crucial solution design characteristics that significantly impact the decision- making process regarding adoption. Recent research has shown that 90% of network outages happen because of human factors and 69% of respondents manage their network with manual method. That high human error of up to 90%, occurs when configuring network devices using manual method. These problems had a negative impact on the functioning of the organization and were harmful to the user. In this research, we investigate solutions to reduce network outage that caused human error using network automation using the Python programming language and multi-threading method. This network automation will reduce configuration time, eliminate human error, and significantly increase efficiency. The aim of this research is to investigate the efficacy of network automation using Python and multi-threading to reduce network outages.The method used in this research is a parallel or multi-tread execution process method using GNS3 as network simulator. Based on experimental results, the multi- thread automation approach is significantly faster than both the serial automation method (67 seconds) and the manual method (248 seconds), this method requiring only 41 seconds to configure all Cisco router devices. If you're looking for speed, look no farther than the multi- thread automation approach, which is 6 times quicker than the manual method and 3.7 times quicker than the serial automation approach. The findings of this study have substantial and immediate implications for the ability of network engineers to speed up the configuration of networks and lower the rate of human error in doing so.

Comparative energy analysis of cooling energy performance between conventional and hybrid air source internet data center cooling system

The widespread need for data processing has led to an increase in internet data centers (IDCs), contributing to the generation of high heat loads and high energy usage. This study discusses a simulation model of a newly proposed cooling system applied to data centers for five selected climatic areas in South Korea. Using experimental data of the newly proposed system, performance correlations are developed by regression analysis method to estimate performance curve coefficients for energy modelling and simulation to predict the annual energy performance of the system and compare the performance with that of a conventional cooling system. The results showed an overall annual average energy efficiency of approximately 14.1% of the hybrid cooling system over the conventional cooling system. The mean seasonal coefficient of performance (SCOP) for the hybrid system and the conventional system is approximately 6.7 and 4.9 respectively for all cases in the winter period while a SCOP of 3.2 and 3.7 respectively for the summer period.

On the use of intelligent metasurfaces in data centers

On-demand wireless links in hybrid data centers can augment the existing network capacity to improve the performance. In the absence of line of sight between the rack transceivers, reflectors have been considered in literature to support high bandwidth, on-demand wireless links within the data centers. Metasurfaces, as a candidate smart reflector technology can be harnessed to bring the intelligence burden to the reflector side which, coupled with the expected low power expenditure, can render the technology a useful alternative. Given the fact that this has not been considered in the literature before, in this paper we highlight the design challenges which emanate from such a consideration and present tools and methodologies which can be utilized to address these challenges, focusing on two problems: wireless link selection to be served by the metasurface, and workload characterization within the metasurface. The former is formulated as an optimization problem with demonstrated effectiveness in minimizing the job completion time. Moreover, the workload analysis reveals a number of interesting attributes, as for example, uneven spatial distribution of traffic on the metasurface's controller network and how this is affected by the metasurface size and location, while the use of beam splitting on the metasurface to support parallel processing and storage functionalities does not significantly affect the incurred workload. Unlike previous work on metasurfaces, which consider mobility-driven reconfigurations, in this paper for the first time we consider data-driven reconfigurations.

A Construction Method of Hybrid Wireless Link Topology VDCell

With the development of cloud computing, the data center network is undergoing great changes. The traditional wired data center network needs a large number of wired links, so it takes huge resources to deploy. For the wireless data center network, although the line installation is simple and the maintenance cost is reduced, the wireless transmission signal is easily blocked. Therefore, this paper proposes a hybrid data center network VDCell based on VLC link, and designs a rule to connect wireless links. Compared with the original DCell topology, it has the following advantages: First of all, the VLC link can be directly installed and used, without centralized control in VDCell, so no additional equipment is needed. Secondly, it is only necessary to add wireless links to the original structure, and the design rules are used to construct the structure, which shortens the communication path length between servers and is very simple to deploy. The increase of data links makes the structure more fault-tolerant. In addition, this paper also proposes a unicast routing algorithm.

Energy saving evaluation of an energy efficient data center using a model-free reinforcement learning approach

To reduce cooling energy consumption, data centers are recommended to raise temperature setpoints of server intake. However, in tropical climates, Data Center operators are still found to be operating at lower temperatures. In this paper, we demonstrate that using a floating setpoint with a lowered temperature value for tropical climates reduces the overall energy consumption of Data Centers as opposed to raising the temperature in a static manner. We achieve this by applying a deep reinforcement learning algorithm to a hybrid data center model that was built from data collected off a highly efficient data center. This generates an optimal control strategy which minimizes the costs of energy consumption while operating within the required set of operational constraints. Following which, we evaluate the behavior of the control strategy to account for the exact sources of energy savings. The deep reinforcement learning algorithm learns by continually interacting with the built Data Center model without any prior knowledge of the Data Center. The algorithm is trained under the full-load and the part-load configuration of the Data Center. Testing results show that further energy savings of up to 3% and 5.5% (under full load and part load respectively) can be achieved with targeted cooling provisioning while operating within constraints in an already cooling-efficient Data Center. We find that while building level optimization studies of Data Centers generally improve energy efficiency, the source of energy savings is not well accounted for. Consequently, our studies show that the reduction of server fan usage and not the reduction of cooling energy consumption is the main contributor of energy savings in a deep reinforcement learning-driven data center operating in the tropics.

Automatic Stream Identification to Improve Flash Endurance in Data Centers

The demand for high performance I/O in Storage-as-a-Service (SaaS) is increasing day by day. To address this demand, NAND Flash-based Solid-state Drives (SSDs) are commonly used in data centers as cache- or top-tiers in the storage rack ascribe to their superior performance compared to traditional hard disk drives (HDDs). Meanwhile, with the capital expenditure of SSDs declining and the storage capacity of SSDs increasing, all-flash data centers are evolving to serve cloud services better than SSD-HDD hybrid data centers. During this transition, the biggest challenge is how to reduce the Write Amplification Factor (WAF) as well as to improve the endurance of SSD since this device has a limited program/erase cycles. A specified case is that storing data with different lifetimes (i.e., I/O streams with similar temporal fetching patterns such as reaccess frequency) in one single SSD can cause high WAF, reduce the endurance, and downgrade the performance of SSDs. Motivated by this, multi-stream SSDs have been developed to enable data with a different lifetime to be stored in different SSD regions. The logic behind this is to reduce the internal movement of data—when garbage collection is triggered, there are high chances of having data blocks with either all the pages being invalid or valid. However, the limitation of this technology is that the system needs to manually assign the same streamID to data with a similar lifetime. Unfortunately, when data arrives, it is not known how important this data is and how long this data will stay unmodified. Moreover, according to our observation, with different definitions of a lifetime (i.e., different calculation formulas based on selected features previously exhibited by data, such as sequentiality, and frequency), streamID identification may have varying impacts on the final WAF of multi-stream SSDs. Thus, in this article, we first develop a portable and adaptable framework to study the impacts of different workload features and their combinations on write amplification. We then propose a feature-based stream identification approach, which automatically co-relates the measurable workload attributes (such as I/O size, I/O rate, and so on.) with high-level workload features (such as frequency, sequentiality, and so on.) and determines a right combination of workload features for assigning streamIDs . Finally, we develop an adaptable stream assignment technique to assign streamID for changing workloads dynamically. Our evaluation results show that our automation approach of stream detection and separation can effectively reduce the WAF by using appropriate features for stream assignment with minimal implementation overhead.

Design and Development of Cloud-Based Cyber Risk Assessment System for a Hybrid Data Center

Design and Development of Cloud-Based Cyber Risk Assessment System for a Hybrid Data Center

(Invited) Medium-Voltage SiC Devices for Next Generation of Power Conversion

As a wide bandgap semiconductor, Silicon Carbide (SiC) power switches and diodes offer the promise of lower conduction and switching losses and more efficient power conversion for a wide range of applications that include but not limited to hybrid vehicles, renewable technologies, industrial motors and data centers. However, in present day, SiC solutions for MV-class (>3.3 kV) unipolar switches and diodes suffer from high conduction losses at elevated temperatures, and SiC bipolar devices such as IGBTs require a prohibitive high forward voltage drop of 3V. Therefore, in these systems, the advantage of SiC technology is diminished compared to the incumbent Si IGBT technology. SiC super-junctions (SJ) can break the unipolar limit and offer an improved trade-off between specific on-resistance and blocking voltage for MV-class applications. To date, multi-epitaxial growth and trench-refill approaches have been demonstrated for 1.2kV and 6.5kV SiC SJ devices. Due to low diffusivity of dopants and shallow projectile depth of implanted atoms in SiC, multi-epitaxial solutions are limited to low voltage switches with thinner drift layers (up to 1200V). Alternately, the trench refill approach generates crystallographic defects that result in leakage at high blocking voltages. Furthermore, achieving uniform, target dopant distribution inside re-grown layers requires complex growth conditions and makes the regrowth process challenging. Recently, SiC charge-balanced (CB) diodes and MOSFETs were reported as an alternative solution to SJ devices with a simpler and scalable fabrication process. In these devices, a novel drift layer architecture is implemented with buried p-doped regions inside the drift layers instead of p-doped pillars. This new approach has been adopted for 3.3kV SiC JBS diodes and 4.5kV SiC MOSFET and results in Ron,sp significantly below the 1-D SiC unipolar limit. In the meantime, a new class of SJ devices is being developed that relies on Megavolt-level implantation to form deep p-type and n-type pillars. This multi-epitaxial SJ approach can be implemented in high voltage SiC devices with a smaller number of regrowth steps. In this work, we review fabrication steps for realizing SiC CB diodes and MOSFETs and report static characterization from room temperature up to 175 °C and dynamic response up to 3kV double pulse switching of such diodes and MOSFETs. Furthermore, we report successful fabrication of 2kV SiC SJ PiN diodes as a first building block of deep-implanted SJ technology and will discuss effect of different activation anneal temperatures on recovering lattice damage following high energy implantation.

A novel hybrid data center architecture employing optically-switched WDM LANs over electrical fat trees

Inter connection network in a data center is the need of the hour as the communication backbone which caters the demands to accommodate a large number of servers with minimum possible end-to-end delay. The traditional Fat tree based topologies play a pivotal role for data center network (DCN) albeit in a low scale, on the other hand the upcoming electrical-cum-optical hybrid architecture demands huge power consumption and exhibit significant end-to-end delay. The present work depicts a proposal of highly scalable novel hybrid architecture employing optically switched WDM LANs (based on ShuffleNet topology) over electrical Fat trees with the use of substantial number of optical devices, the proposed DCN architecture is shown to offer reasonable reduction of end-to-end delay to 12.29 µs for mouse traffic and 10.01 ms for elephant traffic as compared to Optical Switching Architecture (OSA), which has significant 23 ms of delay for any traffic condition.

Designing of a Simulation Tool for the Performance Analysis of Hybrid Data Center Networks (DCNs)

Data center (DC) technology changes the mode of computing. Traditional DCs consist of a single layer and only have Ethernet connections among switches. Those old-fashioned DCs cannot fulfill the high resource demand compared with today’s DCs. The architectural design of the DCs is getting substantial importance and acting as the backbone of the network because of its essential feature of supporting and maintaining the rapidly increasing Internet-based applications which include search engines (e.g., Google and Yandex) and social networking applications (e.g., YouTube, Twitter, and Facebook). Every application has its parameters, like latency and blocking in the DC network. Every data center network (DCN) has its specialized architecture. It has a specific arrangement of layers and switches, which increase or decrease the DC network’s efficiency. We develop a simulation tool that comprises two different DC architectures: basic tree architecture and c-Through architecture. Using this simulation, we analyze the traffic behavior and the performance of the simulated DCN. Our main purpose is to focus on mean waiting time, load, and blocking with respect to the traffic within the DCN.

Fly-path: Traffic-based multi-hop routing approach for hybrid wireless data centers

High data transfer rates achieved by 802.11ad at 60 GHz ISM band enables use of wireless communication in data centers. In this paper, we investigate the possibility of offloading traffic from wired to wireless network in hybrid data centers. By understanding the capabilities of the wireless network, we can design the hybrid data center network accordingly, to achieve better construction and operating efficiency. First, we propose a system model in which each top-of-the-rack switch is equipped with two radios, so that three non-overlapping channels of 802.11ad that are available worldwide can be assigned in an interference-free manner to any configuration of wireless links. Then, we propose multi-hop routing algorithms that assign traffic to wireless infrastructure. These algorithms consist of two families. SP family of algorithms route traffic only over shortest-paths between source and destination pairs. LP algorithms relax this restriction and assign traffic to longer paths when necessary. In order to evaluate the performance of our routing algorithms, we also propose a random data center traffic generation method, based on an analysis of a real-world data center traffic pattern. We evaluate the performance of our allocation methods in terms of different metrics for various network sizes. Results show that our methods can offload significant amount of traffic from wired to wireless network, can achieve quite high throughput, and can utilize wireless links very well.

HeporCloud: An energy and performance efficient resource orchestrator for hybrid heterogeneous cloud computing environments

In major Information Technology (IT) companies such as Google, Rackspace and Amazon Web Services (AWS), virtualisation and containerisation technologies are usually used to execute customers' workloads and applications. The computational resources are provided through large-scale datacenters, which consume substantial amount of energy and have, therefore, ecological impacts. Since long, Google runs users' applications in containers, Rackspace offers bare-metal hardware, whereas AWS runs them either in VMs (EC2), containers (ECS) and/or containers inside VMs (Lambda); therefore, making resource management a tedious activity. The role of a resource management system is of the greatest importance, principally, if IT companies practice various kinds of sand-boxing technologies, for instance, bare-metal, VMs, containers, and/or nested containers in their datacenters (hybrid platforms). The absence of centralised, workload-aware resource managers and consolidation policies produces questions on datacenters energy efficiency, workloads performance, and users' costs. In this paper, we demonstrate, through several experiments, using the Google workload data for 12,583 hosts and approximately one million tasks that belong to four different kinds of workload, the likelihood of: (i) using workload-aware resource managers in hybrid clouds; (ii) achieving energy and cost savings, in heterogeneous hybrid datacenters such that the workload performance is not affected, negatively; and (iii) how various allocation policies, combined with different migration approaches, will impact on datacenter's energy and performance efficiencies. Using plausible assumptions for hybrid datacenters set-up, our empirical evaluation suggests that, for no migration, a single scheduler is at most 16.86% more energy efficient than distributed schedulers. Moreover, when migrations are considered, our resource manager can save up to 45.61% energy and can improve up to 17.9% workload performance.

Optimized wireless channel allocation in hybrid data center network based on IEEE 802.11ad

The massive adoption of Cloud services is entailing a tremendous traffic transiting over data center networks (DCNs). Despite the impressive wired throughput, DCNs are suffering from network congestion. To deal with such resource limitation, first, we put forward a novel Hybrid Data Center Network (HDCN) architecture, based on CISCO’s Massively Scalable Data Center (MSDC) model. HDCN enables high throughput wireless (IEEE 802.11ad) and wired (IEEE 802.3) transmissions. Second, we focus on one-hop inter-rack communications in HDCN and propose a novel wireless channel assignment algorithm. Our goal is to maximize the total throughput over the wireless and/or wired infrastructure. We formulate the problem as a minimum graph coloring. To address this NP-hard problem, we propound two novel i) exact and ii) heuristic wireless channel assignment algorithms in HDCN denoted respectively by GC-HDCN and GH-GC-HDCN. The latter make use of i) column generation and ii) branch and price optimization schemes to resolve the Integer Linear Programming assignment problem while reducing computation time. Based on extensive simulations with QualNet simulator considering the full network protocol stack, the results obtained for both: i) uniform and ii) real Facebook’s workload, show that our proposals outperform the related prominent strategies in terms of latency and throughput.

Combined heat and power generation via hybrid data center cooling‐polymer electrolyte membrane fuel cell system

Combined heat and power generation via hybrid data center cooling‐polymer electrolyte membrane fuel cell system

Joint Workload Distribution and Capacity Augmentation in Hybrid Datacenter Networks

In hybrid datacenter networks, wired connections are augmented with wireless links to facilitate data transfers between racks. The usage of mmWave/FSO wireless links enables dynamic bandwidth/capacity allocation with extremely small reconfiguration delay. Also, on-demand workload distribution, where the workload of a job is divided into multiple tasks that can be distributed/routed to different racks to be processed in parallel, allows better utilization of computational resources in data centers. In prior work, the dynamic wireless capacity augmentation and workload distribution decisions were mostly made independently and in a heuristic manner for serving distributed and parallel computing jobs. In this paper, we propose a novel analytical framework and algorithms to jointly optimize both the wireless capacity augmentation and the workload distribution, to minimize the job completion time. We consider workload that is not amenable to pipelining, fully amenable to pipelining, and partially amenable to pipelining. With extensive simulation studies, we show that the gain (in terms of the reduction in the job completion time) can be very substantial when allowing such joint optimization.

Design and optimization of VLC enabled data center network

Visible-Light Communication (VLC) has the potential to provide dense and fast connectivity at low cost. In this paper we propose SFNet, a novel VLC-enabled hybrid data center network that extends the design of wireless Data Center Networks (DCNs) into three further dimensions: (1) fully wireless at the inter-rack level; (2) no need for a centralized control mechanism on wireless links; and (3) no need for any infrastructure-level alterations to data centers. Previous proposals typically cannot realize these three rationales simultaneously. To achieve this vision, the proposed SFNet augments fat-tree by organizing all racks into a wireless small-world network via VLC links. The use of VLC links eliminates hierarchical switches and cables in the wireless network, and thus reduces hardware investment and maintenance costs. To fully exploit the benefits of the topology of SFNet, we further propose its topology design and optimization method, routing scheme, and online flow scheduling algorithm. Comprehensive experiments indicate that SFNet exhibits good topological properties and network performance.

TIO: A VLC-enabled hybrid data center network architecture

To satisfy the ever-increasing bandwidth demand of modern data centers, researchers have proposed hybrid Data Center Networks (DCNs), which employ high-bandwidth Optical Circuit Switches (OCSs) to compensate for Electrical Packet Switches (EPS). Existing designs, such as Helios and c-Through, mainly focus on reconfiguring optical devices to meet the estimated traffic requirements. However, these designs face two major challenges in their OCS-based networks, namely, the complex control mechanism and cabling problems. To solve these challenges, we propose TIO, a hybrid DCN that employs Visible Light Communication (VLC) instead of wired OCS design to connect racks. TIO integrates the wireless VLC-based Jellyfish and wired EPS-based Fat Tree seamlessly and combines the opposite and complementary characteristics, including wireless VLC direct connection and wired electrical packet switching, random graph, and Clos topology properties. To further exploit the merits of TIO, we design a hybrid routing scheme and congestion-aware flow scheduling method. Comprehensive evaluations indicate that TIO outperforms the Jellyfish and Fat Tree in both topology properties and network performance, and the flow scheduling method also evidently improves performance.

On the Complexity of Non-Segregated Routing in Reconfigurable Data Center Architectures

By enhancing the traditional static network (e.g., based on electric switches) with a dynamic topology (e.g., based on reconfigurable optical switches), emerging reconfigurable data centers introduce unprecedented flexibilities in how networks can be optimized toward the workload they serve. However, such hybrid data centers are currently limited by a restrictive routing policy enforcing artificial segregation: each network flow can only use either the static or the flexible topology, but not a combination of the two. This paper explores the algorithmic problem of supporting more general routing policies, which are not limited by segregation. While the potential benefits of non-segregated routing have been demonstrated in recent work, the underlying algorithmic complexity is not well-understood. We present a range of novel results on the algorithmic complexity of non-segregated routing. In particular, we show that in certain specific scenarios, optimal data center topologies with nonsegregated routing policies can be computed in polynomial-time. In many variants of the problem, however, introducing a more flexible routing comes at a price of complexity: we prove several important variants to be NP-hard.

Modeling and deploying hybrid tenant requests with shared networklets

In recent years, IaaS providers are trying to cater to application layer development, PaaS providers are exposing more infrastructure APIs to customers, and SaaS providers start to allow tenants to build their apps within their software. All of these indicate that, the popular wisdom that cloud computing comes in three flavors—IaaS, PaaS, and SaaS—no longer describes reality. The lines between IaaS, PaaS, and SaaS are becoming blurred as datacenter providers seek to create cloud platforms that can satisfy the needs of enterprises and widen their appeal to developers. In hybrid datacenters, resource requests from tenants are increasingly transforming into hybrid requests that may simultaneously demand IaaS, PaaS, and SaaS resources. This paper tackles the challenge of modeling and deploying hybrid tenant resource requests in datacenter networks, for which we coin “networklet” to represent a set of VMs that collaboratively provide some PaaS or SaaS service. Through extracting networklets from tenant requests and thus sharing them between multiple tenants, we can achieve a win-win situation for datacenter providers and tenants. Extensive evaluations show that, the proposed model and deployment algorithm indeed reduce the completion time of tenant requests while maintaining performance guarantee.