Benefits of Moving to Google Cloud vs. Staying On-Premises

The Google Cloud Platform (GCP) is a leading public cloud service provider launched in April 2008 by Google's parent company, Alphabet Inc. Over the last 14 years, GCP has grown to lead and innovate within the cloud computing industry, specializing in various virtualization engines, creating Kubernetes (K8s), and enhancing data with AI and Machine Learning embedded throughout the platform.

This shift has been fueled by the Covid-19 global pandemic and the transition to home working at scale. However, business leaders have been embarking on cloud-focused digital transformation initiatives that focus on moving away from on-prem solutions to cloud-based infrastructure, and this has been happening in great numbers for at least the last five years.

Cloud computing is growing at an unprecedented rate, with the industry expected to be worth $482 billion in 2022, up nearly 22% from 2021.

What Is On-Premise?

Ever since the computerization of the workplace, traditional IT infrastructure was hosted onsite. Servers and mainframes were hosted in-house with day-to-day operations being conducted by a team of local engineers. Nearly every business would host infrastructure internally, with a typical setup consisting of desktop and laptop computers connecting to an Active Directory realm, perhaps some Exchange Servers, then several business application servers and database platforms.

The larger the company, the more servers had to be managed. Computer rooms would grow into data centers, which in turn would grow into multisite redundant configurations.

In the late 90s, when virtualization first started to get noticed, there was a dynamic shift towards outsourcing IT infrastructure to colocation. Colocation in many ways was the start of private cloud computing as managed service providers started to onboard customers rapidly.

Fast forward to today, as we are in the grip of another major shift in the industry as businesses invest heavily in public cloud infrastructure. But what are the benefits of moving to a public cloud provider such as the Google Cloud Platform? Does this spell the end of on-premise hosting?

To shed light on the cost savings of cloud vs. on-premises servers, SherWeb conducted a study in which it found the average cost of an on-premises server was $1,476.31 per month, while the average cost of a cloud server was $313.90 per month. Resource Availability

Why Choose Google Cloud?

Businesses have already invested significant resources into on-premise infrastructure, local data centers, or colocation facilities. IT systems and business applications are often deeply embedded within the organization, often making the transition to the cloud daunting.

A cloud-first narrative opens the doors to new opportunities, encourages businesses to retain the best employees, and helps to engage the workforce, fostering deeper collaboration in the workplace.

GCP enables cloud computing as an on-demand service, with network access available from anywhere with an internet connection. Private, public, or hybrid cloud solutions can unlock the true potential of your business. Cloud computing is built on elastic technology, meaning that services can scale up, down, in, and out with demand. It creates an agile, flexible, and low-cost infrastructure platform that can grow as your business develops.

Google Scale Networking

The Google Cloud Network is so vast it is candidly referred to as being “Google Scale”, something so big that only Google can provide it. There is merit to this claim because, according to some estimates, Google’s network carries as much as 40% of the world’s internet traffic every day. Google’s network is the largest network of its kind on Earth and is growing rapidly year on year.

Google has invested billions of dollars over the years to build a global infrastructure to power the Google Cloud Platform. It is the Google Network that powers GCP, with a key benefit of the platform being that you can leverage as much or as little as needed.

The majority of Google cloud services are built upon the proprietary technology that drives core Google products such as Google Search, YouTube, Google Maps, and Google AdSense. Additionally, unlike its competitors, Google owns the entire fiber network outright. This is a big deal because GCP is not reliant on third-party network providers’ service level agreements.

At the time of writing, there are currently 29 regions in the GCP Network, housing 88 availability zones and providing 146 network edge locations (points of presence), with access available in over 200 countries.

Google’s network carries as much as 40% of the world’s internet traffic every day.

Security Defined Infrastructure

The security demands put upon modern businesses today are greater than ever before. Security is always high on the agenda in the boardroom, protecting company assets, data, employees, and ensuring compliance and regulatory obligations are met is a significant challenge.

Google Cloud solves many of these problems and has an exemplary reputation with Cloud Security. Cloud providers live or die on their reputation to uphold security best practices.

However, security requires a two-way relationship between the customer and the provider. Google Cloud has created all of the security tools and information needed, but it is the user's responsibility to ensure that security services are implemented correctly. This might include encryption, multi-factor authentication, IAM policies, KMS key implementation, and so on.

GCP is so vast that many cybersecurity threats such as a DDOS attack can simply be absorbed by the size of the infrastructure.

So how does Google Cloud implement security?

It starts with the Intrusion Detection System (IDS), a system that inspects every network packet for spyware, malware, ransomware, and other threats. It's a managed service provided by Google and outputs granular logging that can be ingested into a Security Information and Event Management (SIEM) application and other tools for data analytics.

Another popular tool is Google Cloud Armor, conveniently it can work with on-prem and cloud workloads.

Cloud Armor has built-in features to protect against DDOS attacks. It includes a Web Application Firewall (WAF) that can be intelligently updated to thwart the ever-changing security landscape. A very recent example that demonstrated the power of WAF was automatically protected against the Log4J exploit.

Google Cloud Storage Services are protected by encryption at rest by default, and user identity rules via the Central IAM service with support for physical security keys. Google's hardware infrastructure is protected by the HW design & provenance. The compute engine includes a secure boot stack (cryptographic signatures over BIOS/Bootloader/Base OS images), not to mention all of the on-prem physical security at each regional availability zone.

Resource Availability

It is difficult to drive innovation using yesterday's on-premise technology. So far we have compared on-premise networks and on-premise security. It is clear to see the gulf in each service and how on-premise solutions can struggle to compete.

Transitioning business services and leveraging virtual computing, containerized applications, and serverless infrastructure will continue to drive change.

The ability to leverage almost infinite compute resources in any region is a huge advantage offered by GCP. The ability to deploy a cloud server, a database, a serverless application, or even a vast data lake within a few clicks or by a few lines of code has dynamically changed the IT industry.

The situation on-premise is completely different. The lead time to procure hardware can be weeks or months and is subject to supplier availability. When the kit arrives you need a technical engineer to rack and build the server, ensure there is enough power and cooling, you may even need to wait on a comms provider to enable external connectivity. Not to mention the costs of the hardware, licensing, facilities, monitoring, technical support, maintenance contracts, backups, replication, disaster recovery, and so on.

All of these headaches diminish with Google Cloud. You simply choose the size and performance of the server, cluster, or database, etc., then hit deploy. Solutions can be deployed via Google’s Deployment Manager or any other infrastructure as code tools such as Terraform, Helm (K8s), or Ansible.

The ability to leverage almost infinite compute resources in any region is a huge advantage offered by GCP.

Costs and Financial Accounting

You do have to pay for the network bandwidth, cloud storage, and compute resources once the migration is underway. Be aware that during migration there will be a short period when you are paying for both on-premise and cloud resources, so make it a priority to have a speedy migration.

Did you know you can build and deploy a complex solution with zero upfront costs thanks to the Google Cloud's pay-as-you-go cost model? Additional incentives are given for sustained usage and heavy discounts for 2-5 year commitments.

When businesses become cloud-enabled, costs change for the better. Users have access to a vast range of the latest technology, all available at a fraction of the cost of building similar services in-house. Google Cloud billing introduces predictable and accurate monthly charges, and the biggest advantage is that there are no upfront costs; you only pay for what you use.

The finance teams will enjoy the massive reduction in Capital Expenditure (CapEx) and the introduction of predictable Operational Expenditure (OpEx).

There are a variety of other expenses associated with an on-premise environment like hidden expenses due to how often they are overlooked.

Embrace Open Source

GCP strives to keep the Google Cloud Platform friendly for open source projects and to protect against vendor lock-in. Open source gives businesses the ability to take an existing product and make it work the exact way they want. Open source is free and the idea is that a community of developers worldwide advances the project lifecycle.

Some major projects have specifically been created or sponsored by Google Cloud. This includes Kubernetes, TensorFlow, Istio, Knative, Forseti Security, and Apache Beam. Google is focusing on embedding machine learning and AI technologies into their core GCP products, and it is the open-source projects that are helping to drive this change.

Maintenance and Management

An area that is initially and quite often overlooked is the maintenance and management of existing and cloud compute infrastructure. There is no need for expensive maintenance contracts with hardware vendors, no requirement for in-house 24x7 operations and facilities services, and no upkeep of the digital platform such as expansion, hotfixes, and security updates.

These concerns are now handled completely by Google Cloud. It is the provider's responsibility to ensure capacity is available around the clock and systems are functioning correctly and at peak performance. This generates serious cost savings for business leaders and creates added peace of mind.

There may, on occasion, be old hardware getting retired that may require users to reboot an instance now and then, but Google Cloud's uptime guarantee ensures this is an infrequent experience.

Moving to Google Cloud instantly removes the problem of ensuring the physical infrastructure stack is managed and maintained.

Advantages of Remaining On-Premise?

There are occasions when on-premise may be the better option, and this depends on your business's line of work.

PCI Zoning

PCI Compliance is all about protecting card transactions and, despite nearly all cloud vendors being compliant with PCI-DSS, there is still a reluctance to completely move all card payment services to the cloud.

Legacy Applications

Larger enterprise organizations will nearly always have at least one core business service that runs on a legacy application on a legacy system. It is usually ERP systems and core databases that have been left in limbo because the enterprise is scared of upgrading and breaking the solution.

In these circumstances, on-premise is the way to go. Perhaps consider a hybrid model where compatible and new workloads run in GCP. Google Anthos unifies applications between on-prem, private, and public clouds, and even competitor clouds! Anthos can even be run directly onsite.

Larger Enterprise organizations will nearly always have at least one core business service that runs on a legacy application on a legacy system.

Why Google Cloud?

The benefits of moving to Google Cloud vs. on-premise considerably stack up in favor of GCP. Google Cloud has matured into a leading provider where it's possible to run an application anywhere (Multicloud/Hybrid) and still connect it to GCP services.

Google Cloud is a carbon-neutral business. This is achieved by carbon offsetting, but it is still a remarkable achievement when considering the scale of GCP. That can help to modernize the workplace with machine learning and AI analytics.

If you are just exploring the possibility of introducing Google Cloud services, it's worthwhile to migrate some of the more simple services first such as file storage and simple web applications, then cloudify the more complex services at a later date.

Plan how you will get your business data into Google Cloud. The platform can ingest huge volumes of data over a VPN, dedicated interconnect, or even over the Internet. For organizations with petabytes of data, secure transfer appliances are shipped by Google and transferred to a local point of presence (POP) or similar edge location.

After data loading into GCP, you can reap the benefits of big data and AI/ML services that can extract, transform, and load data into separate GCP services. Learn customer buying habits, track business achievements, and company milestones. Google Cloud provides a faster go-to-market for any business, faster scaling to meet traffic demands for events like Black Friday, all whilst providing a user-friendly and highly secure platform.

Google Cloud provides a faster go-to- market for any business, faster scaling to meet traffic demands for events like Black Friday all whilst providing a user- friendly and highly secure platform.