Eco4Cloud

Consolidate OS in corporate datacenter with E4C (WP)

In this White Paper, we demonstrated the benefits of a consolidation of virtual machines with identical Operating Systems over the same physical hosts. The benefits are both financial and technological. The financial benefits consist of a lower license spending for Operating Systems, and increased control over spending. The main technological benefit is an increased memory sharing among Virtual Machines, which is useful to increase the VM density per physical host.

Avoid Resource Contention with E4C (WP)

A primary Italian Telecommunication company increased its VMware environment performance and automated its management with Eco4Cloud Workload Consolidation and Smart Ballooning. In this white paper we show how our customer automated VMs placement and got rid of memory ballooning and dangerous CPU latency.

Host Thin Provisioning with Eco4Cloud

How to take advantage of Thin Provisioning with the physical hosts of a virtualized Datacenter through Eco4Cloud solutions, provisioning only the number of physical hosts that applications need.

Inter-Cloud Workload Optimization

How to optimize workload distribution, automate workload mobility, save energy costs and reduce carbon emissions in an Inter-Cloud environment.

Memory Efficiency in VMware-Virtualized Data Centers

How to take advantage of Transparent Page Sharing, Memory Limits, Memory Ballooning and Memory Rightsizing in the best way with Eco4Cloud solutions.
Field results show that, even in a small server farm, tens of TeraBytes of RAM can be saved without compromising performances, but actually increasing them.

At A Glance: Eco4Cloud – VMware DRS/DPM Comparison

The document describes the key differentiators of Eco4Cloud suite for Datacenters with reference to VMware’s DRS/DPM: consolidation, scalability, platform independence, inter-Cloud management.

451 MIS Impact Report on Eco4Cloud

451 Research covered Eco4Cloud in a report named “Eco4Cloud: Consolidate virtual machines to accumulate datacenter savings”, covering relevant market context, technology, business model, competition and SWOT analysis.

Highly reputed 451 Research is a division of The 451 Group, which provides syndicated research, data and advisory services focused on the business of enterprise IT innovation, critical and timely insight into the competitive dynamics of innovation in emerging technology segments such as cloud computing, datacenter technologies and Internet infrastructure services.

Saving energy in datacenters through workload consolidation

This whitepaper on recently co‑authored with 4 top-notch European excellence centers such as the Institute for High Performance Computing and Networking of the Italian National Research Council, the Department of Electronics and Telecommunications at Politecnico di Torino, eERG – Energy Department at Politecnico di Milano and PrimeEnergyIT/EfficientDataCenters frames the whole workload consolidation topic and provides an overview of state-of-the-art approaches, including E4C’s of course.

Eco4Cloud and Cisco UCS: Unified Efficiency

This whitepaper reports the outcomes of a Proof of Concept test carried out at one of Cisco IVT Lab, in San Francisco, between March 28th and April 2th, 2014. The resulting CapEx and OpEx reductions are tangible so as the advantages deriving from the integration of Eco4Cloud with Cisco Unified Computing System (UCS).

Saving energy in datacenters with Eco4Cloud and Cisco UCS

This white paper frames the workload consolidation topic and provides an overview of state-of-the-art approaches, including Eco4Cloud. Moreover, the document clarifies how Eco4Cloud can be profitably integrated with Cisco Unified Computing System (UCS). Indeed, the abstraction of resources offered by Cisco UCS is the perfect starting point for the workload consolidation provided by Eco4Cloud: UCS creates a homogeneous environment tailored to applications migration and dynamic server hibernation and resume.

Eco4Cloud’s PoV on Micro-Servers Energy Efficiency

This document describes the PoV of Eco4Cloud in the context of low-power processor micro-servers, whose adoption is increasing in these years. The consolidation algorithm of Eco4Cloud helps to achieve significant reductions of consumed energy, between 30% and 60%.

Workload Consolidation

Eco4Cloud offers a software platform which minimizes the energy consumption of virtualized data centers by dynamically consolidating Virtual Machines on the minimum number of physical resources.

This document lists targets and features of Eco4Cloud Virtualization Infrastructure Management solution.

Smart Ballooning

Smart Ballooning enables the release of memory which is actually not used by VMs and make it available to ESX, which will possibly allocate it to other VMs in demand.

Capacity Decision Support Manager

Capacity Decision Support Manager provides a solution to efficiently plan and use shared pools of resources in virtualized data centers within few hours from the initial deployment.

Troubleshooter

Eco4Cloud’s Troubleshooter automates monitoring of the virtualization platform, notification of underlying misconfigurations, with impact analysis, related severity ranking and solutions recommendation.

Architecture and Requirements

Eco4Cloud is an unobtrusive Virtualization Infrastructure Management software for virtualized datacenters. It abstracts from hardware and OS levels and interacts only with virtualization layer.

This documents describes the software architecture, setup phases, and technology requirements.

Avoid Resource Contention with E4C

This document unveils the performance improvement and IT management automation benefits brought by Eco4Cloud in a primary TelCo deployment.

Telecom Italia Case Study

This document reviews Eco4Cloud Workload Consolidation deployment in Telecom Italia Information Technology and related energy consumption and costs reduction.

PoC Report at Herning Kommune

This document reports the results achieved on the server farm of the Danish Municipality of Herning Kommune, with both Workload Consolidation and Smart Ballooning enabled.

Use Case

This document describes the results obtained with a server farm from a Telco company. It reports why and how the customer obtained a 50% energy savings and measure the yearly cost savings.

ROI modeling in a standard Eco4Cloud deployment

Eco4Cloud uses a three-step methodology for conducting ROI analysis:

1. Gather quantitative benefit information. In this study, the benefits amount to energy savings in a data center, while costs amount to the E4C license purchase and renewal.
2. Create a complete investment (three-year total cost analysis) profile based on costs and benefits.
3. Calculate the ROI and payback period. E4C conducts a depreciated cash flow analysis of the investments over a three-year period.

The final results on a mid-range data center are 111% ROI, with 7 months of payback period.

Managed Service Provider

A large Managed Service Provider has selected Eco4Cloud’s technology to deliver significant cost savings in their highly virtualized and multi­tenanted cloud environment. Eco4Cloud actively manages the number of powered­on cores in the server estate resulting in significant reductions in electricity costs (30%) due to reduced server power draw and reduced cooling demands. In addition they have been able to deliver Server Thin Provisioning, provisioning only the number of physical servers that their customers use, rather than the number that they request. This feature is common in storage environments but is unique to Eco4Cloud in the virtualization world.
The sum total of benefits include: reduced data center space, reduced numbers of physical servers, reduced power and cooling demands all of which lead to better ROI/profitability and a more competitive offer to clients.

Large Enterprise

A large enterprise firm has selected Eco4Cloud to manage their test and development environments. Eco4Cloud have been able to manage a constantly changing workload where different environments are in use at different times, without over provisioning the physical server estate or consuming unnecessarily large quantities of electricity for servers that are doing no practical work. All of this is done seamlessly and automatically with no administrator intervention.
Eco4Cloud’s Smart Ballooning memory technology enables memory to be actively managed and put where it is needed as workload moves depending on test patterns and schedules. Eco4Cloud also manages the active cores that the whole workload needs to deliver service, ensuring that sufficient cores are always available but never more than are needed consuming cash and other resources.
Benefits include: optimization of RAM and CapEx deferral/reduction for the purchase of additional capacity, OpEx reduction due to energy savings >30%, increased health posture of the virtualization infrastructure as crucial parameters such as CPU Ready-Time and Memory Ballooning have dramatically improved.

Outsourcing

An outsourcing firm has selected Eco4Cloud to provide assistance in migrating workload from Windows 2003 to Windows 2014. Whilst the underlying server estate was largely suitable for the upgrade there was insufficient headroom to manage the swing of services from one operating system to the other. Eco4Cloud shrunk the existing estate to make that headroom and provide free physical servers to support the migration.
Benefits include: business agility due to reduction of migration period, CapEx savings as no additional hardware was required to support the migrations operations, higher computing density at final steady­state.

User Manual

This document, the Eco4Cloud User Manual, describes how to install and use the Eco4Cloud Virtualization Infrastructure Management software components, monitor and manage the information generated by the components, build your Eco4Cloud datacenter environment, consolidate virtual machines within the datacenter environment, and set up rules for virtual machines, hosts and clusters included into the datacenter environment.

Architecture and Requirements

Eco4Cloud is an unobtrusive Virtualization Infrastructure Management software for virtualized datacenters. It abstracts from hardware and OS levels and interacts only with virtualization layer.

This documents describes the software architecture, setup phases, and technology requirements.

E4C PoC Requirements

Some simple technical requirements are needed in order to run a Proof of Concept installation of Eco4Cloud’s products. These documents relate to VMware vSphere and Microsoft Hyper-V environments respectively.

Scientific Overview

The Eco4Cloud project originates from years of research activities of a group of four researchers of the ICAR-CNR Institute of the National Research Council of Italy. The four researchers – Agostino Forestiero, Raffaele Giordanelli, Carlo Mastroianni, and Giuseppe Papuzzo – are among the co-founders of the company, later recognized as a spin-off from the CNR and the University of Calabria. The core product, the software for the dynamic consolidation of workloads in data centers, founds on an innovative decentralized/probabilistic algorithm based on the swarm intelligence paradigm and on bio-inspired techniques. The expertise and skills of the CNR research group allowed the very complex consolidation problem (known to be NP-hard, i.e. exponential, when tackled in a classical centralized/deterministic fashion) to be solved efficiently even in large data centers.

These issues are covered in the following scientific overview.

White Paper

This white paper, co‑authored with 3 top-notch European excellence centers such as the Department of Electronics and Telecommunications at Politecnico di Torino, eERG – Energy Department at Politecnico di Milano and PrimeEnergyIT/EfficientDataCenters, frames the whole topic of workload consolidation with a small yet comprehensive amount of scientific details.

IEEE Transactions on Cloud Computing

The scientific paper “Probabilistic Consolidation of Virtual Machines in Self-Organizing Cloud Data Centers”, published in late 2013 on IEEE Computer Society – Transactions on Cloud Computing, is the most complete scientific work discussing the mathematical principles of the Eco4Cloud’s algorithm as well as the results of Proofs of Concepts performed in real data centers. The paper was selected as the “flag paper” for the first semester of 2014 by IEEE TCC, which is recognized as one of the most prestigious  journals in this field.

Inter-Cloud Workload Optimization

The following white paper describes how EcoInterCloud extends the basic EcoCloud consolidation solution to the “Inter-Cloud” scenario in which several data centers or cloud instances are interconnected, and applications can be routed or migrated among them to optimize workload distribution based on policies. The paper introduces several business goals that can be achieved with EcoInterCloud and includes a high-level technical description of the algorithms used for the assignment and migration of Virtual Machines in an Inter-Cloud environment.

Publications

More scientific papers, reported in the list below, were published in the last years in the proceedings of prestigious international conferences. They cope with different and important aspects of the consolidation problem. For each work, references are provided to the paper and/or the slides of the talk and/or the presented poster.

For more information please contact the Scientific Director of the company, Carlo Mastroianni.