Workload profile · Gallery websites
Web hosting for art galleries — The archive is the site; the private view is the spike
Most weeks nobody reads your 2019 exhibition page. On the evening a mailing list goes out, several thousand people open the same three URLs at once.
The short answer
A gallery site has two entirely different load profiles and they need sizing separately. The archive is hundreds of rarely-visited pages, where what matters is whether a cold request is quick. The private view is a concentrated spike on a handful of URLs, where what matters is whether the cache is warm before the email goes out.
Both are handled by NVMe behind a LiteSpeed cache, which answers a cached page from server memory without starting PHP. Sprint costs $2.42 a month with free SSL, free migration and daily backups; Nitro at $7.08 adds two cores, 4 GB of RAM and 200 GB.
By the HostingFast team · Reviewed 24 August 2026
24/7
Support answering, day or night
1-click
WordPress deployed in a click
Free
SSL certificates, no charge
Daily
Backups taken daily on all plans
The current exhibition, an archive of past shows, the artists you represent, and visitor information. Four sections, and one of them grows forever while the other three are rewritten every few weeks.
That structure produces a distinctive measurement problem. An archive page is almost never in the cache when somebody finally opens it, so the honest number for the archive is the uncached response time. The current exhibition page is the opposite: it will be requested thousands of times in an evening and should be served from memory.
The sections below treat the two separately, then deal with the send itself — because a mailing list going out to a few thousand recipients is the real load test a gallery site faces.
Two traffic patterns, one account
Measure the archive with the cache deliberately missed — append a query string, or use a page nobody has opened in a month — and time the response. That is what a researcher following a link from a catalogue actually experiences, and NVMe storage is what makes it acceptable.
Then measure the current exhibition page normally. It should be answered from the LiteSpeed cache, which means PHP never runs and the database is never touched. The difference between those two figures tells you exactly how much work the cache is doing for you.
An archive of past shows is a cache-hit problem
Hundreds of low-traffic pages cannot all be held in cache at once, and there is no point pretending otherwise. What you can do is make the uncached render cheap: keep archive templates simple, avoid plugins that query every post on every page load, and keep images at display size.
Redis object caching, available on the managed WordPress plans, helps here in a different way from page caching — it holds the results of database queries, so a page that must be rendered is rendered faster. That is the right tool for an archive, where full page caching cannot help.
The mailing list send is your real load test
A private view email lands in several thousand inboxes at once and produces a sharp, narrow spike on two or three URLs. Warm the cache before you send by loading those pages yourself, so the first recipient is not the one paying to render them.
Send the mail properly too: outbound goes through MailChannels here, with SPF, DKIM and DMARC configured on the domain, which is what determines whether an invitation reaches the inbox at all. A beautifully fast page nobody was told about is not an achievement.
Sizing against the exhibition calendar
For most galleries Sprint at $2.42 a month is genuinely sufficient: one site, 10 GB, 500 GB of bandwidth, and a cache that absorbs the spikes. The reason to move up is a shop, a ticketed event, or a second site for a project space.
Nitro at $7.08 brings 200 GB, two cores, 4 GB of RAM, backups every six hours and Imunify360 in front of the account. For a site holding an archive that represents years of institutional memory, the tighter backup cadence is the part worth paying for.
Quick when it is cold, calm when it is busy
NVMe storage keeps the uncached archive page acceptable and the LiteSpeed cache keeps the current exhibition page instant. Those are two different jobs and the platform does both without configuration.
Free SSL, free migration and daily backups come with every plan, with the cadence tightening to every six hours on Nitro and Imunify360 in front of the account.
- NVMe storage for the archive page nobody has cached
- Server-level LiteSpeed cache for the exhibition everyone opens
- Backups every six hours on Nitro, daily on every tier
- MailChannels on outbound mail, with authentication set up
Why HostingFast
Standard on every plan
Cold pages that still land
NVMe storage means the archive page a researcher opens after two years does not feel like an archive.
Spikes absorbed in memory
A cached exhibition page is returned without PHP starting, so a mailing list send is uneventful.
Invitations that arrive
MailChannels handles outbound mail with SPF, DKIM and DMARC configured on your domain.
Backups for institutional memory
Daily on every plan, every six hours on Nitro, with Imunify360 in front of the account.
Room for a project space
Turbo carries 25 sites and Nitro 100, each with its own domain and certificate, on one bill.
Humans on the evening of a private view
Support is staffed at every hour and aims to reply within one when a site is down.
Quick Start
From order to online
- 1
Time a cold archive page
Load a page nobody has opened recently, or append a query string, and record the response time as your archive baseline.
- 2
Move the site in free
Our engineers migrate at no charge and you sign the result off before DNS changes; a fresh build installs in one click.
- 3
Warm the cache, then send
Open the exhibition pages yourself before the mailing list goes out, so the first recipient gets a cached response.
Built In
Loaded onto every plan
- Uncached response time measured on an old archive page
- LiteSpeed page caching for the current exhibition URLs
- Redis object caching available on the managed WordPress plans
- NVMe SSD storage on every tier, including the entry plan
- MailChannels on outbound mail, with SPF, DKIM and DMARC set
- Backups every six hours on Nitro, daily on the other tiers
- Imunify360 in front of the account on the Nitro tier
- Free SSL on every plan, renewed automatically before it can expire
- Softaculous installs 240+ applications in one click
- Human support on duty every hour of every day
Frequently Asked
What people ask us most often
Why does an old exhibition page feel slower than the current one?
Because it is almost certainly not in the cache. A page-level cache holds what is being requested; a show from four years ago is not. When somebody finally opens it, WordPress renders it from scratch, which means PHP runs and the database is queried. NVMe storage is what keeps that acceptable, and Redis object caching on the managed WordPress plans makes the render itself quicker.
How do I stop a private view email overwhelming the site?
Warm the cache before you send. Open every URL the email links to yourself, so the pages are already in the LiteSpeed cache when the first recipient clicks. After that the spike costs almost nothing, because a cached page is answered from server memory without PHP starting. The traffic is entirely logged out, which is the easiest kind there is.
Is page caching or object caching the right tool for an archive?
Object caching, mostly. Page caching only helps a page that is requested repeatedly, and archive pages are not. Object caching stores the results of database queries, so a page that must be rendered from scratch is rendered faster — which is exactly the archive's problem. Redis object caching is available on the managed WordPress plans for that reason.
What backup cadence should a gallery archive have?
Daily is the standard on every plan, with a one-click restore you run yourself. If the archive represents years of institutional record, the every-six-hours cadence on Nitro at $7.08 is the part of that tier worth paying for, and Imunify360 sits in front of the account as well. Losing a day of an archive is a very different loss from losing a day of a brochure site.
Keep reading
Recruitment Agencies — Hosting Guide
Listing pages that grow forever, and the caching strategy that keeps the index quick.
Moving Companies — Hosting Guide
Sites with a seasonal spike and a long quiet middle, sized for both ends.
Uptime (Glossary)
What an uptime figure is actually measuring, and what a credit clause is worth in practice.
Drupal Hosting — Built for Art Galleries
Running the archive on Drupal instead, with the same measurements applied.
Domain Names
Registering and pointing a name, with the timings that decide when a change goes live.
Changing hosts? Run through our checklist first.
A straightforward sequence for a switch your visitors never feel: which files move first, how to shift email across without losing a single message, the right moment to repoint DNS, and the two mistakes behind almost all the downtime we get asked to rescue.
Size it for both patterns.
NVMe for the cold archive, LiteSpeed for the spike, from $2.42 a month with free SSL, migration and daily backups.
View Drupal Hosting — Built for Art Galleries plans