Actions

Cloud Bursting

Cloud Bursting is a technique used by hybrid clouds to provide additional resources to private clouds on an as-needed basis.[1]

In cloud computing, cloud bursting is a configuration that’s set up between a private cloud and a public cloud to deal with peaks in IT demand. If an organization using a private cloud reaches 100 percent of its resource capacity, the overflow traffic is directed to a public cloud so there’s no interruption of services.[2]


Cloud Bursting
source: WordPress


The technology behind cloud bursting includes infrastructure as a service (IaaS), a cloud computing infrastructure that enables developers to dynamically configure cloud resources on demand. DevOps tools like configuration management and infrastructure as code (IaC) are used to specify the capacity thresholds or trigger events that will activate cloud bursting.


When is Cloud Bursting Needed[3]

Cloud bursts are triggered by spikes in computational resource demand. These bursts could be an influx of user traffic or expensive one-off computational tasks. Some common cloud bursting situations include:

  • Software Development Software development and analytics are two of the most common use cases for cloud bursting. DevOps teams often use multiple virtual machines for testing purposes, which are only needed for a short amount of time. Plus, CI/CD tasks are good candidates for bursting into the public cloud, since CI/CD requires several one-off tasks that run for a short amount of time when developers push new commits.
  • Marketing campaigns: Marketing campaigns for new product launches can generate a huge influx of traffic that requires extended cloud resources. Imagine the marketing push for an anticipated Hollywood movie or video game release. These events generate a temporary stampede of traffic that subsides after the launch news passes, so it is a great time to utilize cloud bursting.
  • Big data modeling and queries: Big data companies frequently need to execute one-time queries or generate models that will exceed the capacity of their private cloud. These tasks are well suited to cloud bursting. The company can burst into the public cloud for additional resources to expedite the task. Some granular examples of big data tasks include:
    • High-fidelity 3D rendering
    • AI and ML model training
    • Autonomous vehicle simulation
  • Seasonal businesses: Seasonal businesses need additional computational resources during known peak times. For example:
    • Holiday rush shopping for an eCommerce or shipping site
    • End-of-business-quarter financial processing
    • Political election seasons (campaign fundraising and website traffic for education on a candidate's proposals)

In conclusion... When businesses need agility to react quickly to changing business demands that strain cloud computing resources, they can burst into the hybrid cloud. They expand cloud capacity for a short (or long) time, without needing to expand private cloud resources indefinitely.


Considerations When Using Cloud Bursting[4]

Cloud bursting can offer many advantages. However, there are some important considerations to keep in mind before implementing it.

  • Data transfer charges: First, it’s important to be aware of any potential data transfer charges between public and private clouds. Cloud providers are known to charge hefty fees for transferring data — which can drive up the cost of a cloud bursting project to the point where it’s cheaper to buy private computing resources instead.
  • High latency: Cloud bursting can also create latency — especially when using cloud storage systems located far away from the data center. Bursting certain resources to the public cloud can cause applications to suffer from performance issues. One way to reduce issues is to use edge resources that are located close to users.
  • Potential security issues: Generally speaking, businesses should be cautious about bursting sensitive data to the cloud as it requires running data across multiple locations. Businesses are strongly encouraged to back up data when bursting into the cloud to avoid any potential mishaps. It’s also important to extend authentication tools to the public cloud to ensure that applications remain protected.


Cloud Bursting Scenario[5]

The Figure below illustrates a Cloud bursting scenario. In our example, we have an existing local network which is composed of a ProActive Scheduler with a Resource Manager. This resource manager handles computing resources such as desktop machines or clusters. In our figure, these are referenced as local computing nodes. A common kind of application for a scheduler is a bag of independent tasks (no communication between the tasks). The scheduler will retrieve through the resource manager a set of free computing nodes to run pending tasks. This local network is protected from Internet with a firewall which filters connections.


Cloud Bursting Example
source: Brian Amedro, Denis Caromel et. al


When the scheduler experiences an uncommon load, the resource manager can acquire new computing nodes from the Amazon EC2 Cloud. This decision is based on a scheduling loading policy, which takes into account the current load of the scheduler and the Service Level Agreement provided. These parameters are directly set in the resource manager administrator interface. However, when offloading tasks to the Cloud, we have to pay attention to the boot delay which implies a waiting time of few minutes between a node request and its availability for the scheduler. However, there are approaches by which the Resource Manager is capable to bypass firewalls and private networks.


IT Infrastructure Needed for Cloud Bursting[6]

Typical cloud bursting scenarios include spikes in seasonal e-commerce traffic, season finales of popular television series that break viewership records, or intense intervals of application development and testing. As user demand increases, private on-premises or cloud infrastructures need to quickly and seamlessly move either full or partial instances of applications and data to a duplicate infrastructure in the public cloud. For cloud bursting to be successful, there are a number of issues to consider, including:

  • Replication: The state of applications and data when replicated in a hybrid cloud environment must remain in synch so nothing gets lost during the cloud burst. Seamless orchestration (i.e., containers) and automation between private and public cloud environments is needed to provide a consistent application development environment, app/database synchronization, workload balancing and app failover.
  • Network: Legacy networks were not designed with agility in mind let alone cloud bursting, where at a moment’s notice, high-bandwidth, low-latency connectivity is required to meet the volitivity in workload demand.
  • Security: Direct access to public cloud infrastructures may not be local to centralized data center applications and data, requiring workloads to transverse over wide area networks (WANs) or the public internet, impeding performance and increasing the risk of outages and cyber-attacks.

For applications to quickly access large volumes of public cloud compute or storage resources, or share data/app access across a hybrid cloud infrastructure, you need direct and secure interconnection that delivers low network latency and high-throughput at the edge, where the most user demand is being generated.


Importance and Benefits of Cloud Bursting[7]

Cloud bursting is all about dynamic deployment of applications that normally run on a private cloud into a public cloud to meet expanding capacity requirements and handle peak demands when private cloud resources are insufficient. Cloud bursting can make these private clouds more cost-efficient by eliminating the need to overbuild physical infrastructure to ensure enough capacity to meet fluctuating peaks in demand. Private clouds can be rightsized in terms of compute and storage to accommodate the ongoing demands, because the peaks can be handled by a public cloud and a pay-per-use model.

There are many scenarios where businesses can benefit from cloud bursting. For example, many sectors deal with seasonal spikes that put an extra burden on private clouds. Enterprise data centers may have geographic needs where one location experiences heavy loads and must meet application-specific performance needs.

Software development projects and analytics are two of the fastest-growing drivers of demand for cloud bursting. DevOps teams spin up numerous virtual machines for testing purposes that are only needed for a short time. For example, organizations in the health and financial sectors are heavy users of HPC and often have fluctuating analytics needs that require high-core-count applications such as:

  • Computationally intensive, read-heavy workloads
  • Batch or burst usage model for longer-running simulations
  • Risk analysis (credit, hedge funds, portfolio management, actuarial)
  • Log analysis

Overall, the availability of a public cloud offers the chance to reduce the capital cost of owning and managing excess compute capacity and storage for all kinds of workloads. By combining it with on-premises cloud resources and using cloud bursting to manage it, the public cloud serves as on-demand, overflow capacity and eliminates the need for costly over provisioning to meet temporary demand.

Hybrid clouds are particularly useful in certain sectors. Life sciences workflows, for example, generate a great deal of data on premises for things like genomic sequencing but rely on the ability to analyze and compute the data in the cloud. This is a prime example of a temporary-use scenario. Cloud bursting is also integral to the financial sectors, which must develop predictive models based on stock market data for market risk analysis.

The ability to effectively and seamlessly manage demand and potentially bring thousands of additional cores to bear is the only way that many sectors can make this type of data analysis effective in terms of both cost and time. When demand spikes, these companies can do the cost/schedule math to figure out how much additional processing power is needed and just rent it from Amazon Web Services or Microsoft Azure.

To seamlessly manage and monitor cloud bursting of HPC workloads in a hybrid cloud requires a sophisticated cluster management solution that can integrate a wide variety of workload managers and eliminate the high learning curve across cloud platforms. This automation software solution can integrate across all workload management solutions by significantly reducing the complexity inherent to cloud bursting. This solution brings a great deal of agility, responsiveness, and simplification that saves money and time while opening up cloud compute vistas for enterprises across every sector.


See Also

Cloud Computing
Cloud Computing Reference Architecture (CCRA)
Cloud Integration
Cloud Manufacturing
Cloud Migration
Cloud Service Architecture (CSA)
Private Cloud
Public Cloud
Hybrid Cloud
Hybrid Cloud Computing


References

  1. Defining Cloud Bursting Intricately
  2. What Does Cloud Bursting Mean? Microsoft
  3. When do organizations need cloud bursting? Atlassian
  4. Key Considerations When Using Cloud Bursting Datameer
  5. Example of a Cloud bursting scenario Brian Amedro, Denis Caromel et. al
  6. Is your IT infrastructure built to burst? Equinix
  7. Why is Cloud Bursting Important? Bright Computing