AWS offers guidance for trusted cloud connections

As written by Stephanie Kanowitz on gcn.com
A new resource is available to help agencies develop Trusted Internet Connections (TIC) architectures in the cloud.

So far, the capabilities of TIC, an Office of Management and Budget mandate to reduce the number of network gateways on federal networks and route external connections through approved government agencies -- TIC Access Providers or Managed Trusted Internet Protocol Services -- are not available in the cloud.

But Amazon Web Services’ “Guidance for TIC Readiness on AWS,” released Feb. 3, details ways that agencies could develop TIC-ready architectures on the AWS cloud, rather than routing traffic through a TICAP or MTIPS, which can slow connections and cause constraints on a the network. The guidance, based on results of a pilot program, addresses how agencies can directly access applications running in a TIC-ready cloud on a Federal Risk and Authorization Management Program moderate baseline.

The guidance highlights two areas: common connection scenarios with a TIC overlay and AWS capabilities and features that help with TIC compliance. The scenarios include a use case involving authenticated web and mobile applications in an “all in cloud” deployment, such as the General Services Administration’s GSA Advantage, which is a public website with authentication requirements.

“In this architecture, an [Internet gateway] provides Internet connectivity to two or more customer-defined public subnets across multiple Availability Zones in the [virtual private cloud],” the guidance states. “An [Elastic Load Balancing] load balancer is placed in these public subnets. A web-tier is configured within an Auto Scaling group, leveraging the ELB load balancer to provide a continuously available web front end. This web tier securely communicates with other backend resources, most notably the backend identity store used for role-based authentication.”

Another scenario involves public web and mobile applications requiring authentication and operating in hybrid environments. This means a portion of the environment is situated onsite in a data center. Users can access these applications from home or via public Wi-Fi or agency networks using either personal or government-issued devices.

In this case, part of the application architecture resides in the cloud while the other -- often sensitive-data sources -- reside in a data center. “Connectivity between the in-cloud portions of the application and the controlled, on-premises components is achieved using AWS Direct Connect or virtual-private network service in conjunction with a TICAP or Managed Trusted IP Service provider,” the guidance states. “In this way, data flow between the customer’s in-cloud and on-premises services are seen by the TIC.”

The second aspect of the guide discusses the capabilities and features available to achieve TIC compliance in the cloud. It includes AWS Identity and Access Management, which is a web service that enables IT departments to manage multiple users, groups, roles and permissions for AWS offerings such as the Amazon Relational Database Service.

Amazon CloudWatch is another on the guide’s list. It’s a monitoring service for AWS cloud resources and the applications that run on them. It can collect and track metrics, monitor log files and set alarms, providing systemwide visibility into resource use, application performance and operational health, the guidance states.

Other capabilities and features include:

• Amazon Simple Storage Service, a scalable distributed object store that stores objects redundantly on multiple devices and at multiple facilities.

• Amazon Elastic Compute Cloud, a web service that enables resizable compute capacity in the cloud.

• Amazon Config, a managed service that provides an AWS inventory and configuration history and sends configuration change notices.

Amazon issued the guidance after completing the testing phase of the FedRAMP-TIC Overlay pilot. The program started in May 2015 as a way to research an approach that would address agencies’ need for fast and secure connections. Currently, mobile users connect to an agency, which connects to a FedRAMP-approved cloud provider via TICAPS or MTIPS. In the future, mobile users would connect with a FedRAMP-approved cloud that is also TIC-compliant, and that cloud would then connect with an agency via the trusted providers.

Amazon worked with Homeland Security Department and FedRAMP officials on the testing.

AWS OpsWorks for Windows now supports custom AutoScaling based on Amazon CloudWatch Alarms and custom AMIs.

Amazon CloudWatch alarms can be used as thresholds for AWS OpsWorks Automatic Load-based Scaling. For example, you can use ‘Disk Reads’ or ‘Network In’ as metrics to scale up or down your load-based instances.

Custom AMI support gives you the ability to use your own AMIs based on a Windows Server 2012 R2 base that have software such as Microsoft Windows Server 2012 R2 with SQL Server Express, SQL Server Standard or SQL Server Web preinstalled.

These additions to AWS OpsWorks Windows support benefit customers that want to start with custom built AMIs and customers that have the need to scale their infrastructure based on any metric available in Amazon CloudWatch alarms.

Read our Product Page to learn more and our Documentation to get started with AWS OpsWorks.

Source: http://aws.amazon.com/about-aws/whats-new/2015/08/aws-opsworks-for-windows

By Jeff Barr, Amazon WorkSpaces

AWS customers are deploying Amazon WorkSpaces at scale in medium and large organizations. For example, health care company Johnson & Johnson is using WorkSpaces to realize the long-promised security and efficacy benefits of virtual desktops, in a world populated by a diverse workforce that would like to use their own computing devices if possible (also known as BYOD – Bring Your Own Device). You can view their recent presentation, Deploying Amazon WorkSpaces at Scale, to learn more about what they did and how they now support BYOD for 16,000 contractors and employees, along with zero clients for another 8,000 users.

New Metrics

In order to help our customers to monitor their WorkSpaces deployments, we recently added additional Amazon CloudWatch metrics for WorkSpaces. These metrics are designed to provide administrators with additional insight in to the overall health and connection status of individual WorkSpaces and of all of the WorkSpaces that belong to a particular directory.

Like all CloudWatch metrics, these metrics can be viewed in the AWS Management Console, accessed via the CloudWatch APIs, and monitored by CloudWatch Alarms and third-party tools.

The new metrics are enabled by default and are available to you at no extra charge.

Here’s what you get:

• Available – WorkSpaces that respond to a status check are counted in this metric.

• Unhealthy – WorkSpaces that do not respond to the same status check are counted in this metric.

• ConnectionAttempt – The number of connection attempts made to a WorkSpace.

• ConnectionSuccess – The number of successful connection attempts.

• ConnectionFailure – The number of unsuccessful connection attempts.

• SessionLaunchTime – The amount of time taken to initiate a session, as measured by the WorkSpaces client.

• InSessionLatency – The round trip time between the WorkSpaces client and WorkSpaces, as measured and reported by the client.

• SessionDisconnect – The number of user initiated and automatically closed sessions.

Here’s how you create an alarm that will fire if a user cannot connect to their WorkSpace:


Available Now: The new metrics are available now and you can start monitoring them today!

Source: https://aws.amazon.com/blogs/aws/new-cloudwatch-metrics-for-amazon-workspaces/