MediaAgility, in Partnership with Google Cloud is deploying Rapid Response Virtual Agents to help you address the overwhelming demands in your contact centers.
These virtual agents run on Google’s Contact Center AI technology that is powered by Google Dialogflow.
In order to serve our customers during this unprecedented situation that we’re in right now due to the outbreak, we’re offering free deployment of these virtual agents for all of our customers till July 31st, 2020. The virtual agents can be deployed in less than 2 weeks.
They come loaded with a lot of Coronavirus related data from Center of Disease Control and Prevention (CDC) and can answer custom questions from your bank of Frequently Asked Questions (FAQs).
We are deploying these agents for free and Google is not going to charge for the platform till July 31st 2020 either. So this is absolutely at no cost to you and can help you address your customers’ needs.
Optimizing cloud consumption and reducing the money spent on IT infrastructure is one of the strategic priorities of any CIO
Here are the 8 secrets that you can leverage to reduce your cloud costs.
Secret #1. Custom Machine Types
Instead of provisioning a virtual machine (VM) with one of the many available standard configurations, you can actually create a custom configuration. Instead of overprovisioning, you can carefully assess exactly how much compute power you need for your workload and resize to the desired configuration.
Secret #2. Rightsizing Recommendations
This is related to the first secret. Google Cloud actually monitors your resource consumption and recommends you to rightsize your VMs to save cost.
And when you click on the recommendation, you can resize your VM with the click of a button.
Secret #3. Per Second Billing
This is less of a secret if you are already on the cloud, but if you are running your infrastructure on-premise, this is a huge benefit why you might want to consider moving to the cloud.
Google cloud charges all services on a per-second basis. So if you use a VM for 90 seconds, you pay for 90 seconds.
Imagine adding an additional 64-core server to your existing on-premise datacenter that you need only for a few minutes per day to handle peak load, or to run your batch jobs, etc. With Google Cloud, instead of paying for that infrastructure up-front, you consume only for the time when you actually use the servers. You move from a CapEx model to an OpEx model.
This means that the large upfront cash that you now save can be applied towards innovation and growth.
Secret #4. Preemptible VMs
This is clearly my favorite secret from this list. Google Cloud leases out its excess capacity for much cheaper, up to 80% cheaper!!
There are a few things to note with Preemptible VM
They can last maximum of up to 24 hours
Google can pull the plug on (preempt) these VMs with a 30-second warning
They are easy to set up. Just set the “Preemptibility” to “on” when creating a VM.
Preemptible VMs are best suited for batch jobs or fault-tolerant jobs.
Secret #5. Cloud TPU
Cloud TPU is a custom chip built by Google for running Deep Learning workloads. TPUs can save you both time and cost while training very deep neural networks.
Looking at this experiment by Martin Gorner, it is clear that a TPU can help you not only speed up your model training but cut your model training costs into half.
Secret #6. Sustained Use Discount
This is a discount that Google provides to you by default. The longer you leave a VM running, the bigger the discount is applied. You can’t do anything to accidentally forget to get the discount.
On average, customers save 21% off of list price by doing nothing.
Secret #7. Committed Use Discount
You can commit to a certain amount of compute and memory consumption to get up to 57% discounts in many cases.
You don’t have to commit to any particular machine type, all you need to commit to is the number of Cores and the amount of Memory consumption and then you can consume those resources with any combination of any machine types.
Secret #8. Offline Commit Contracts
If you are a large corporate or an enterprise, you can get into a multi-year commit contract with Google Cloud to get an additional 5% to 15% discount. This discount is applied on top of all other discounts and savings that we’ve discussed here.
The easiest way to leverage this secret and execute a commit-contract is to work with a Google Cloud Premier Partner like MediaAgility.
Connect with me on LinkedIn if you’d like to learn more or would like to schedule an assessment of your environment to understand how much can you save on your cloud costs.
To start with, let’s determine how much it’ll cost us to provision a Jupyter notebook environment on AI Platform notebooks with a similar configuration as that of a Colab. Navigate to Google Cloud Platform > AI Platform > Notebooks > New Instance.
Select n1-highmem-2 machine type, which has a similar config as Colab. i.e. 2 cores and 13GB RAM. Also, select the Tesla K80 GPU, which is the most economical GPU on GCP.
When you select this configuration, on the right hand of the screen you’ll see that Google estimates that this instance will cost you $293.25 per month.
An hourly rate of $0.402 is determined after applying the sustained use discount, which means if the Machine is not used continuously throughout the month, the hourly rate might go up. This clearly shows the value that Colab provides by giving you an environment for free.
Having said above, Google Cloud AI Platform Notebook does have its own advantages. For example, it gives you an environment that is integrated with the rest of the Google Cloud Platform and can be managed by your IT team.
Here is a side by side comparison of the two products:
AI Platform Notebooks
2 cores, 13 GB RAM (double of that in Colab Pro), 1 GPU
Any size you want. Scale on-demand
Resides on Google Drive
Part of the Google Cloud Platform. Easy GIT Integration
Maximum runtime of 12 hours. (24 hours for Colab Pro)
No limitation on runtime. Pay-per-use with per-second billing
IT generally has no visibility
Part of GCP, hence IT has full control and can manage notebooks similar to all other resources on the cloud
Easy authentication to Google Services. Pre-built connectors for Google Sheets, Google Cloud Storage, etc.
Fully integrated with Google Cloud Platform. Easy access to data on Google Cloud Storage, BigQuery, etc.
Consumer product, no enterprise support
Part of the Google Cloud Platform. Enterprise support to resolve all your issues and queries
Gotta pay for this guy
Good for personal research, science projects
Good for building a managed experimentation environment for enterprise ML systems
To conclude, whether you select Colab or AI Platform Notebooks, really depends on your use case. If you want to use a notebook for personal research or some science projects, use Colab. If you are a Data Scientist and want to use notebooks for building enterprise ML systems, then use Google Cloud AI Platform Notebooks.
People often ask me why Google Cloud specifically for Analytics & AI? By the way of this post, I am trying to answer why I choose Google Cloud to build Analytics & AI systems.
If you look at Analytics or Machine Learning, the core building block for Analytics & AI is Data. You can not build a Machine Learning system without having a ton of data and without having a platform that can process that amount of data.
Google’s mission is “to organize the world’s information and make it universally accessible and useful”. Google is a data company.
Data comes from the consumption of services. There are only 13 services in the world that have at least a Billion Monthly Active Users users.
Two of them are by Microsoft, i.e. Windows and Office; Two by Facebook, i.e. Facebook and Whatsapp; And nine of them by Google.
All of you are probably familiar with each one of them and use them on a daily basis.
Google is a unique company with 9 of these apps with more than Billion Monthly Active Users, many of them with more than two billion users. Just think about how much data these apps must be generating.
All of these services are free for you. This data is critical to monetization of these apps. These apps are highly data-intensive and are fully loaded with Machine Learning.
This means Google needs to build products and infrastructure that can process and analyze that amount of data.
Which means they have the capability, the infrastructure, engineers, and algorithms to be able to run Analytics & Machine Learning at that scale.
Google has built Machine Learning systems for close to 2 decades now. Because of the amount of data that Google generates and processes, it gives me confidence that there is no other company in the world that runs Analytics and Machine Learning at the scale that Google does.
My name is Arpit Agrawal. I am the Director of Analytics & AI at MediaAgility and work closely with Google Cloud to solve some real-world problems for corporate and enterprise businesses. I focus on leveraging Google Cloud to generate insights from data. Insights that help you make intelligent decisions that propel your growth.
Through GCP.LIVE, I hope to bring you my decade worth of experience working on various data analytics & AI projects.
Please feel free to reach out if you find the content useful, or have feedback on anything that I write. Also, do let me know if there are additional topics that you’d like me to cover.
Looking forward to partnering with you in your journey towards building an intelligent enterprise.