BYTE the Cloud

Chris 0:00
Hey everyone and welcome back for another deep dive. Always

Kelly 0:03
great to be back. Today,

Chris 0:04
we're going to be talking about the AWS snow family of services. Sounds good as mid level cloud engineers, you're probably pretty familiar with AWS, but yeah, this is a service that you may not have a lot of experience with, right? So we're gonna do a deep dive into it. Okay? We're gonna talk about what it is. We're gonna talk about why it's important. Okay, we're gonna talk about some of the use cases, and we're also gonna cover how it might appear on the solutions, architect, Associate exam. Cool.

I'm excited to dive in. All right, so to get started, what is the AWS snow family?

Kelly 0:39
So the AWS snow family is a collection of services and physical devices that are all about moving lots of data to and from AWS, okay, so

Chris 0:49
physically moving data? Yeah, exactly. So we're talking like hard drives in the mail. Yeah. Pretty much seems a little old fashioned in the age of cloud computing, it

Kelly 0:56
does, but it makes sense for certain situations. Think about when you have massive amounts of data, except like petabytes or even exabytes, trying to transfer all that data over the Internet, especially if you have a slow or unreliable connection, would take forever, right? Snow family offers a secure way to move that data offline. So it's kind

Chris 1:15
of like the heavy duty truck of the data world,

Kelly 1:17
exactly. So

Chris 1:18
when would someone actually use this in the real world,

Kelly 1:21
there are tons of cool use cases. Let's say you're a media company and you have a huge library of video footage that you need to get into the cloud. Okay, yeah, we're talking hundreds of terabytes, maybe even petabytes, of data. Shipping it physically, using snow family would be way faster than trying to upload

Chris 1:38
it all. That makes sense. What are some other examples. Another

Kelly 1:40
great example is scientific research. Let's say you have a team of researchers working in a remote location like Antarctica. Oh, wow, where the internet is super limited, they can use snow family to get their data back to their labs safely and reliably.

Chris 1:55
So snow family isn't meant to replace our everyday Data Transfer Tools, right? But it's more of a specialized tool for these unique situations, exactly,

Kelly 2:04
and knowing when to use snow family is really important, especially if you're preparing for the AWS exam.

Chris 2:10
Okay, so we've talked about what snow family is and why it's important, but let's dive into the specifics. Sounds good. Are we talking about just one service here, or are the different options?

Kelly 2:20
There are actually a few different options within the snow family. Each one is designed for different data transfer needs. Okay,

Chris 2:26
so it really is a family of devices. Exactly.

Kelly 2:28
You've got Snowcone, Snowball and Snowmobile. Each one has its own special features and capacity. Okay,

Chris 2:35
I'm definitely intrigued. Now you too.

Kelly 2:37
Let's start with Snowcone. It's the smallest member of the family. It's designed to transfer up to eight terabytes of data. It's small and easy to carry around,

Chris 2:46
so it's kind of like the carry on suitcase of data transfer. Yeah, good way to put it. What about those bigger jobs?

Kelly 2:52
For those you have scobal, it's a really tough device that can handle up to 80 terabytes.

Chris 2:57
Wow, that's quite a jump up from snow con it is. Think of Snowball

Kelly 3:01
as a super durable shipping container built to survive even the harshest conditions.

Chris 3:06
So Snowcone for the smaller stuff, Snowball for the big leagues. Is that it? Or is there something even bigger? Well,

Kelly 3:13
if you need to transfer absolutely massive amounts of data, there's Snowmobile, okay, and get this, it's a truck sized data center. It can transfer up to 100 petabytes a truck full of data. You got it. We're talking exabytes here. It's designed for those absolutely massive data transfers,

Chris 3:30
like moving an entire data center to the cloud. Exactly.

Kelly 3:33
It's the ultimate moving van for your digital life.

Chris 3:36
Okay, snow family is way more interesting than I initially thought. Yeah, it's pretty cool. But besides the devices themselves, what are the benefits of using snow family

Kelly 3:46
security is a major one. These devices are tamper resistant, okay? And they have built in encryption and chain of custody tracking, so your data is protected every step of the way. So

Chris 3:56
it's not just about the physical transfer, but also about keeping that data safe and secure Absolutely.

Kelly 4:01
And another big benefit is that snow family integrates seamlessly with other AWS services. Once your data gets to AWS, you can use it with S3, EC2, IAM, all that good

Chris 4:13
stuff. That makes a lot of sense. So we're not just shipping data off into a black hole. It's ready to be used within our AWS environment. Exactly.

Kelly 4:20
It's all part of a unified experience. But it's important to remember that snow family isn't a perfect solution for every situation.

Chris 4:28
Oh, okay, are there any limitations we should be aware of? Yeah, for

Kelly 4:32
sure, snow family is not a good fit for real time applications or for small data transfers,

Chris 4:37
so no live streaming video games using Snowmobile, probably not

Kelly 4:42
the best use case. Snow family is all about moving large amounts of data offline, securely and reliably for smaller real time stuff. The internet is usually a better option. All right. I

Chris 4:51
think I have a good understanding of the different snow family devices and their purpose. Great. But let's shift gears a little bit now and focus on how all of this relates to the eight. US Solutions Architect Associate exam, what kind of questions should we be prepared for? Definitely

Kelly 5:04
expect scenario based questions like you might get a scenario where a company needs to transfer 50 terabytes of data from a place with limited internet access. Okay? And the question would be, which snow family device would you recommend?

Chris 5:21
So understanding the capabilities of each device is crucial.

Kelly 5:25
It is you need to know which device is the right fit for the situation.

Chris 5:29
Got it what other types of questions might we see? You

Kelly 5:33
might also get questions about the technical details of snow family, like the security features, how it integrates with other services, things like that. So it's

Chris 5:41
not just about knowing that snow family exists, but understanding how it works

Kelly 5:46
exactly. And don't forget about cost optimization. You might need to compare the cost of using snow family to other methods, right? We need to

Chris 5:53
think like Cloud architects and find the most efficient solution exactly. Okay? I think it would be super helpful if we could go through some example exam questions and really break them down. Yeah, let's

Kelly 6:02
do it. That's a great idea. So

Chris 6:03
let's jump into some examples. What do you have for us?

Kelly 6:06
All right, let's say you have a research team in the Amazon rainforest, and they need to transfer 100 terabytes of ecological data back to their headquarters. Wow. Okay. Internet connectivity is unreliable and super expensive there, which AWS snow family device is the most suitable solution? Okay,

Chris 6:24
so we have a lot of data, 100 terabytes, and the internet's not really an option, exactly. So that rules out our typical data transfer methods, yeah. So if we look at our snow family options, Snowcone is definitely too small, yeah, way too small. That can only handle eight terabytes, yep. And Snowmobile, which is that truck size data center, right? That's definitely overkill, way overkill. It can handle 100 petabytes, exactly. So that leaves us with Snowball, yes, which can handle up to 80 terabytes. And

Kelly 6:53
it's built for those tough conditions. Yeah, perfect for the rainforest,

Chris 6:55
exactly. But I know the exam loves to throw in some tricky options, so why wouldn't we choose something like AWS Direct Connect or S3 transfer acceleration in this scenario? Good

Kelly 7:07
question. AWS Direct Connect is a dedicated network connection between your on premises environment and AWS. It offers high bandwidth, but it requires a reliable internet connection, which we don't have in this case,

Chris 7:20
right? And Direct Connect isn't really designed for offline data transfer anyway, exactly.

Kelly 7:23
What about S3 transfer acceleration? Do you remember what that does? Yeah, that

Chris 7:28
speeds up uploads to Amazon S3 but it still relies on the internet, right? It does,

Kelly 7:32
which is a problem in this scenario. Plus S3 transfer acceleration is really meant for smaller, more frequent transfers, not one massive transfer like we have here.

Chris 7:42
Okay, so we really have to consider all the details in this scenario to figure out the best answer, exactly. Okay, let's try a more technical question, Okay, how

Kelly 7:49
about this? Which of the following security features is not offered by AWS snow family? Okay? A, tamper resistant devices. B, built in encryption. C, chain of custody tracking, or D integration with AWS Shield.

Chris 8:05
Okay, let's see the devices themselves are designed to prevent tampering. So Option A is definitely a feature. And we talked about how the data on snow family devices is encrypted automatically, both at rest and in transit. So option B is in two, yes. And there's chain of custody tracking, which tells us where the data is every step of the way. So option C is also a feature. You got it. So that leaves us with Option D integration with AWS Shield. And what is AWS Shield? AWS Shield is a DDoS protection service. It

Kelly 8:34
is and it's great for security, but it's not directly related to snow family. So the answer is, D You got it. AWS Shield is important, but it's not a feature of the snow family itself. Okay, that

Chris 8:45
makes sense. Any other tips for tackling these exam questions? Pay attention

Kelly 8:49
to keywords. Look for clues that point you to the right service or feature good

Chris 8:54
advice, okay, hit me with another scenario. I'm ready. Okay, how

Kelly 8:57
about this? A genomics company needs to archive two petabytes of raw sequencing data for long term storage, cost efficiency is a major concern. Which AWS storage solution is most appropriate?

Chris 9:10
Okay, so massive data set, two petabytes long term storage and cost efficiency, right? What jumps out at you? Amazon S3 standard is great for data we need to access frequently, but probably not the best for archival. Definitely not. And EBITDA volumes are meant for EC2 instances, yep. So it doesn't really fit this scenario, either. Nope.

Kelly 9:29
What about S3 Glacier deep archive? Any thoughts on that one? Ooh.

Chris 9:34
S3 Glacier deep archive isn't that specifically designed for long term archival, and it's super cost effective. You got it. So that's our answer that yes,

Kelly 9:43
S3 Glacier deep archive is the perfect fit for this scenario. Okay, but

Chris 9:47
what about regular S3 Glacier, or S3 one zone in frequent access? Why wouldn't those work?

Kelly 9:53
Regular S3 Glacier has higher retrieval fees than Glacier deep archive. Oh, okay, and S3 one zone, infrequent. Access is for data you access less frequently, but not necessarily data you're storing for years and years. So it's

Chris 10:06
all about choosing the most cost effective option for this specific use case. Exactly. Okay,

Kelly 10:11
I'm getting the hang of this great ready for another one?

Chris 10:13
Absolutely. Bring it on.

Kelly 10:15
Okay, let's test your knowledge of Snowball Edge. A company is using Snowball Edge to transfer data to AWS, which of the following is a valid use case for Snowball Edge's local compute capabilities.

Chris 10:28
Ooh, Snowball Edge now we're talking about edge computing, so it's not just about transferring data, but also processing it on the device itself. Right? Okay, so what are our options?

Kelly 10:40
You could run machine learning models to analyze the data before sending it to AWS.

Chris 10:45
Ooh, that's a good one. Yeah, that's

Kelly 10:47
a great way to save time and bandwidth, right? What else? You could also host a static website that can be accessed locally, even without an internet connection,

Chris 10:54
so even if you're in a remote area, you could still access critical information. That's really cool, it

Kelly 10:59
is. But remember, not all use cases are a good fit for Snowball Edge,

Chris 11:03
okay, like, what

Kelly 11:04
like performing real time video transcoding for live streaming events. Hmm,

Chris 11:08
yeah, that feels a little off. I mean, Snowball Edge is powerful, but it is powerful. They're really designed for real time processing. Good question,

Kelly 11:16
and the answer is no, Snowball Edge has those local compute capabilities, but it's not meant for real time applications that need super low latency. Something like real time video transcoding would be better handled by a service like AWS elemental media live. So for

Chris 11:33
Snowball Edge, we're talking more about pre processing data or maybe hosting local websites Exactly.

Kelly 11:38
It's not a replacement for services specifically designed for real time stuff. I'm

Chris 11:43
feeling pretty good about Snowball Edge. Now. We've covered a lot of ground here. Yeah, is there anything else we should touch on before we wrap things up? Definitely

Kelly 11:50
make sure you understand how snow family works with AWS security services like IAM and KMS, right? We talked

Chris 11:56
about snow family's own security features, but it also needs to fit into our overall AWS security posture Exactly. So how does it work with IAM and KMS?

Kelly 12:05
IAM controls who has access to your snow family devices and the data on them, okay? And KMS lets you manage the encryption keys for your data. So we're

Chris 12:15
not just treating snow family as this separate thing. It's all integrated Exactly.

Kelly 12:19
And don't forget about cost optimization. Make sure you understand how new family pricing works, including data transfer costs, storage fees, device rental, all of that.

Chris 12:29
Got it think like a cloud architect. Be cost conscious exactly any final exam. Tips.

Kelly 12:35
One last thing I'd say is to stay up to date with the latest documentation, because things in AWS are always changing, right? You know, don't be afraid to get some hands on experience with snow family. It really helps to solidify your understanding. Yeah,

Chris 12:49
absolutely. I'm a big believer in learning by doing. Me too.

Kelly 12:52
All right. Are you ready for one last exam style question before we wrap up? Absolutely. Bring it on. Okay? This one focuses on security. A company needs to transfer highly sensitive data to AWS using Snowball which of the following security measures should they implement? A, use a strong password to protect the Snowball. Device B, encrypt the data before transferring it to the Snowball. C, enable multi factor authentication for all users accessing the data. Or D implement a Virtual Private Network or VPN connection for data transfer. Okay,

Chris 13:26
lots to think about here. Security is so important, especially when we're dealing with sensitive information, it is so let's go through the options. Option A using a strong password to protect the Snowball device. Okay, does Snowball even use passwords?

Kelly 13:42
That's a good question, and the answer is no, okay, Snowball doesn't use passwords. It uses a job manifest that authorizes the data transfer and has encryption details. So

Chris 13:52
Option A is out. It is, what about option B, encrypting the data before putting it on Snowball? That's a

Kelly 13:59
great idea. Yeah, it's always best practice to encrypt sensitive data before putting it on any storage device. But it's not strictly required with Snowball.

Chris 14:06
Oh, right, because Snowball already has built in encryption. It does.

Kelly 14:10
It encrypts the data at rest and in transit, so encrypting beforehand

Chris 14:14
just adds an extra layer of security. Exactly. Okay, Option C enabling multi factor authentication for all users accessing the data. That's important for overall security, but it is. Is it relevant to Snowball itself?

Kelly 14:27
That's a good point. Multi factor authentication is more about controlling access to AWS services and accounts, not for physically interacting with the device.

Chris 14:37
Okay, so probably not. The best answer here, what about option D, implementing a VPN connection for data transfer.

Kelly 14:44
So here's the thing, about Snowball the data transfer doesn't happen over the public Internet, right? You physically ship the device to AWS, and they take care of importing the data. So a VPN isn't really necessary, exactly. It's all done offline. And insecurely. Okay, so the best answer here is B Yes. Encrypting the data before transferring it to Snowball is a great security practice, even though Snowball itself already handles the encryption. Awesome.

Chris 15:10
I feel like I'm really starting to understand the nuances of snow family. That's

Kelly 15:14
great to hear. You're well on your way to acing that exam. You know, it's

Chris 15:17
funny, as mid level cloud engineers, we tend to focus on all those everyday AWS services, but snow family is proof that sometimes the best solution isn't the most obvious one. It's like having this secret weapon in our toolkit

Kelly 15:29
exactly, and knowing when to use that secret weapon is what's gonna set you apart. Remember, those exam questions are all about choosing the right tool for the job,

Chris 15:39
and sometimes the right tool is a giant truck

Kelly 15:43
full of data. Exactly. It's pretty amazing to think about, yeah,

Chris 15:47
in the age of lightning fast internet, you wouldn't think we'd still be physically shipping data around. You wouldn't, but for those massive data sets in those hard to reach places, snow, family is a game changer. Absolutely.

Kelly 15:58
Think about working on a project with putta bytes of data, like you're analyzing satellite imagery, you're not gonna transfer that over the internet, right? It would take way too long, and it'd be super expensive.

Chris 16:10
Yes, snow family opens up a ton of possibilities. It's perfect for things like genomics, research, media production, all those data intensive fields, totally. And even though we're talking about physical devices. Snow family is still a part of that AWS ecosystem. It is. So once the data gets to AWS, we can use it with S3 EC2, machine learning, all that good stuff, exactly. It's all connected. It's pretty cool how AWS blends cutting edge tech with these practical solutions. It's awesome. And even though snow family seems complicated at first. It does. It's actually pretty straightforward. Once you break it down, it

Kelly 16:44
is and remember the why is just as important as the what, absolutely,

Chris 16:47
you know, this got me thinking about how I could use snow family in my own projects. Oh, yeah. What if we combined snow wall Edge's local compute capabilities with serverless functions to process data right at the edge.

Kelly 17:02
Now you're thinking like a true cloud architect, that's

Chris 17:04
the kind of creative problem solving we need.

Kelly 17:06
Absolutely it's what's gonna make you stand out in the field. Well, I

Chris 17:10
think we've covered pretty much everything there is to know about snow family. I think so too. Thank you so much for all the insights. I feel way more confident about those exam questions now. Awesome.

Kelly 17:18
Glad to hear it

Chris 17:20
and for all our listeners out there, yeah, keep studying, stay curious, and don't be afraid to experiment. There's always something new to learn in the cloud. So true. Thanks again for joining us for another deep dive till next time. Happy clouding.