Voice of the DBA

Tag: hardware

  • Expensive CPUs

    There have been a lot of features added to the SQL Server platform over the years. Several of these features let us perform functions that are beyond what a database has traditionally been designed to handle. SQL Server has had the ability to send emails,  execute Python/R/etc. code, and in SQL Server 2025, we can call REST endpoints.

    Quite a few of these features (arguably) are more application-oriented than database-oriented. There’s nothing inherently wrong with having a server perform some of these functions, and there have been some very creative implementations using these features. I recently ran into one of these examples from Amy Abel, where she shows how to use the new REST endpoint feature to call an AI LLM to generate and send emails from your database server. That’s creative, and it’s reminiscent of the numerous examples from various experts over the years who demonstrate how these features can be used to accomplish a task.

    However, these are examples. They work amazingly well with one user running a limited workload. This reminds me of many of the examples I’ve seen using the AI vector enhancement in SQL Server 2025 to create embeddings from string data using an LLM. That is interesting, but most of the examples show a trigger being used to update the encodings. Imagine users updating data and those triggers firing. Imagine a real workload and how often your users might update string data you want to use in an AI application, especially a RAG application. Think about how complex or long-running triggers in your applications now that can overload your system.

    CPUs in database servers are expensive. The hardware isn’t more expensive, but the software is pricey. Standard Edition is limited to 24 or 32 cores (depending on version), and while Enterprise isn’t limited to any number of cores, the cost of each core is $$$$. Is it worth having those $$$$ cores sending emails or calling external services? Or would you be better offloading those calls to another server, like an app server, where the cost of the core is the hardware and a little .NET code running separately?

    Many of us already struggle with the database server as a bottleneck for our application and workload. Scaling up our database systems is expensive and cumbersome. We struggle to get approval for larger VMs, and if we scale up in the cloud, it gets very expensive very fast. I’m not surprised that database vendors are happy to add these features as it increases the licensing cost for applications using them.

    I know the majority of the cost of building applications is labor and software developers’ time. However, that’s changing with LLMs that can produce code cheaply. I keep seeing that the cost of writing code is approaching zero.

    That’s not going to be true if you use LLMs trained on the example code that increases your database licensing cost. Then the cost might be higher than you expect.

    Steve Jones

    Listen to the podcast at Libsyn, Spotify, or iTunes.

    Note, podcasts are only available for a limited time online.

  • Building a Data Center

    Most of you reading this have likely seen a data center, and maybe sat in one while connecting to a database, but how many of you have had to try and do some of the work to set one up? It’s not something most data professionals do, but since we often have some of the most expensive boxes in one, I thought it would be interesting to discuss what goes into getting one running.

    There’s a blog post that starts to cover what it takes to set up a data center. I found it interesting to read as I’ve been a part of the group choosing a space a few times, and a lot of the issues discussed in the blog were things I dealt with or experienced. Both with large and small companies, I’ve helped to spec out and decide what we need from a colocation facility or our own room in an office. In smaller companies, I’ve had some of the “what-the-duck” moments outlined in the post when I tried to rack and connect systems, only to find some issue with power, connections, etc.

    When I read that post, I think about the months of getting everything set up, it’s hard to see this as more cost-effective than the cloud in some ways. Certainly setting up racks and servers is easy. Getting them powered, having a cold and hot aisle, and deciding what gets connected to where these are relatively simple things. However, making decisions on networking, routing, and other complex tasks can take specialized expertise. For a single company, finding, hiring, and retaining someone who knows this stuff well is hard. Startups might get someone excited, but the really good people who can design this stuff are often working at a cloud company of some sort.

    Maybe those people are tired of the FAANG/Microsoft world and want to come to your startup, but for how long? And how do you replace them? Lots of companies found maintaining a data center to be more expensive than it was worth.

    I suppose a similar situation in the database world is setting up a complex system, perhaps something like a new Firebolt warehouse. If the person that did that leaves, can the rest of the company fill their shoes? Do you know how Firebolt works? Would it be better to use a more popular system, like Snowflake or Databricks, with lots of other people writing articles and blogs about how things work? I think it would be, if for no other reason than there are other models to follow and more people that can give you guidance.

    A data center, whether a private one or a cell in a large cloud provider’s footprint, is a complex beast. The days of using a closet with servers on a bookshelf are likely gone for most commercial enterprises. Even SQL Server Central moved on to hosted systems. We used to rent a rack in the early 2000s in a co-location facility, one I enjoyed visiting, as long as the visits were few and far between. Now I like connecting to EC2, which is much simpler and easier.

    Steve Jones

    Listen to the podcast at Libsyn, Spotify, or iTunes.

    Note, podcasts are only available for a limited time online.

  • The Vast Expansions of Hardware

    At the Small Data conference recently, one of the talks looked at hardware advances. It was interesting to see a data perspective on hardware changes, as many of us only worry about the results of hardware: can I get my data quickly? In or out, most of us are more often worried about performance than specs. However, today I thought it might be fun to look at a few changes and numbers to get an idea of how our hardware has changed, in the march towards dealing with more and more data. Big data anyone?

    In thinking about disks, I saw a chart that looked at the changes from HDD (hard disk drives) to SDD (solid state drives) to NVMe (Nonvolatile Memory Express). These show read speeds going through the list from 80MB/S to 200MB/s to 5000+MB/s. That’s a dramatic change, and not one only in high-end arrays. There are off-the-shelf drives you can put in a desktop that read this fast. If you think about some of the early IBM drives, which read at 8800b/s. Growth in disk speed, inside the timeline of our careers, has grown by a few orders of magnitude in read speed.

    Write speed hasn’t grown as much but capacity has. My early career work used HDDs with a 100MB capacity. These days we can get TB range storage on all of these mediums, with many laptops having 0.5TB or more on them. Desktops often have plenty more. My current workstation at home has 3.5TB of storage. Contrast that to the early IBM drive linked above, which had 5MB. These days people regularly demo hundreds of TB, or even 1PB queries from a database.

    Many of us just expect the network to work well. In fact, I assume many of us won’t complain to network people since they are never at fault for performance issues. I started my career with Arcnet connections between machines. Those ran at 2.5Mb/s. We were moving those and 4Mbps Token ring to Ethernet at 10Mbps with Thicknet, Thinnet, and eventually RJ-45 connections. When we got 100Mpbs bridges, I thought we were cutting edge for our SQL Server Central servers. If we look back 20 years, 1Gbps was more the standard then, but today we see growth up into the 800Gbps with Infiniband. While I don’t know many data centers doing that, there are plenty running in the 50Gbps range.

    If we think about CPUs, I started my career on a 386 machine running at 25MHz. I helped upgrade some 286 machines, but most of our servers were 486 class machines at 25 MHz. I still remember being excited about the early Pentium processors for a large system. There were many Pentium variants and later families of processors, but back in the 2000s, almost all machines were single-core. The first multi-core chips were released and slowly became more common over time. These days, many new laptops have multiple cores, including the new on I got, which has 12 cores. If you want, you can purchase an AMD Epyc 9004 processor with 96 cores. That’s on one chip. Since most servers can take more than one CPU, you can have hundreds of cores running if you want. If you want to get really crazy. the Nvidia Blackwell has thousands of cores for their GPU-based AI calculations.

    Memory has likewise grown, though it seems most servers are much less than a TB of RAM, which is a much lower growth over time than storage and networking. Maybe because of those two changes, memory has had less of a reason to grow into common multi-TB-sized capacities in our systems. In fact, for you reading this, what are the common memory sizes you have in servers? I see many VMs and other machines set up with somewhere between 128GB and 1TB for memory, even as their data sizes have grown much, much larger. However, there are plenty that don’t have anything near 128GB.

    That was one of the interesting things I realized about the Small Data conference, and one reason the event was created. Most of our data sets, especially usable sets, and most of our queries can run on a laptop if not a mobile device. The focus on big data seems overblown, especially as most of our companies don’t have anything approaching 100TB, much less 1PB. If you need it, there is hardware out there for you, but some of the amazing advances made over time are lost on me as the common, average capabilities out there on the majority of systems could handle the majority of my needs.

    With some well-written queries.

    Steve Jones

    Listen to the podcast at Libsyn, Spotify, or iTunes.

    Note, podcasts are only available for a limited time online.

  • One Month with the Dell Latitude 7450

    It was just over a month ago that I got a Dell Latitude 7450 from our corporate IT group. It wasn’t my first choice, but as Redgate grows, they’re trying to be more secure and standard. I didn’t have a good reason to not try it, to I agreed to give it a try. My boss said if it was a problem, we’d get an exception and I’d choose another one. I had hoped for another HP Spectre, but this really is just a toaster to me. I need it to just work.

    The Design

    This is a bit heaver than my old Spectre, but also larger. It also has more ports.

    I got the laptop, so it doesn’t fold over into a book or tablet or tent. I had thought that might be useful at times, but in 8 years of two Spectre versions that opened past 10 degrees, I think I’ve done that 5 times.

    The screen is nice and works well for me. I like that I have both USB C and USB B ports on the laptop. I wish I had one of each on each side, but I have 2 Cs on the left and 2 Bs on the right. It’s a very minor gripe.

    I do like I have an HDMI port built in, which is something my Spectre was missing. This means I can take something else out of my travel kit.

    I got this with 32GB of RAM, which is probably a good reason why Windows 11 runs smoothly here. I also got a 1TB drive, which is overkill. I had a 512 in the last machine and after 5 years, I still had space.

    The keyboard is OK, fairly standard, backlit, and responds well. The trackpad is large, and seems to work well with the left/right clicks.

    The touchscreen is good and it also works with a pen, and I got an Active one from Amazon that I’ve used in whiteboard sessions with customers.

    Annoyances

    The big thing was the trackpad for me. It is a little sensitive and if a finder or hand brushes it, I get some weird actions. Mostly I do something the causes windows to minimize, which is really annoying. I’ve tried tuning down some of the actions, but I need to do more. It doesn’t cause too many problems, but enough that it’s on my mind.

    The Home and End keys moved to the top row. They were on the side of my old machine, and I was used to those. I still can’t quite hit home and end without looking, or without hitting insert. I also lost my page up/down keys as well. I can use Funtion+up arrow (or down), but that’s not something I think about.

    Concerns

    I’ve had 2 blue screens coming out of a closed machine in a month. The system was unresponsive once and when I tried to CTRL+ALT+DEL it blue screened after a few minutes.

    I’m keeping an eye on it. This might be my machine or this model, but if I get a few more, I might need to replace it.

    Overall

    Overall I learned that I don’t really care anymore about a laptop. It’s a tool, and I need it to work, but I’m not that wedded to the hardware. They keyboard matters to me, but of a bunch I tested, they were all find enough. Small differences, but overall they worked.

    Other than that, this thing works well and runs fine. I still think some of the Win11 stuff that changed wasn’t an improvement and I need to figure out if I can turn some things off, but it works well.

    I’d get another one of these without complaints. Assuming the blue screens stop.