PHPMem v2.0.1
Version
1.6.45
Uptime
16 days 21 hours 8 minutes 2 seconds
Memory
Total
512MB
Used
12,72MB (2.48%)
Free
499,28MB
Keys
Current
14 060
Total (since start)
40 994
Evictions
0
Reclaimed
760
Expired Unfetched
0
Evicted Unfetched
0
Connections
Current
15 / 1 024 max
Total
192 387
Rejected
0
llm:12634c98b7b8c09ab1eac55622474faa4eca219e6ec7f725889c54edecfcd341
Edit
Yes. **WEBSITE** is the natural grouping, and **YEAR** is a useful second axis. The table's grain is one row per website, year and page, and the 11 websites differ sharply in size, engagement and lifespan (from the per-website profile I ran).
**1. Traffic scale (by WEBSITE)**
- **Dominant properties:** `city.brla.gov` has about 720.8M page views from only 404 page rows. `www.brla.gov` has about 237.2M views from 24,597 rows. Together they account for nearly all traffic.
- **Mid-tier, many pages:** `brgov.com` has 15.2M views over 17,595 rows, and `data.brla.gov` has 12.5M over 14,379 rows. Both are long-tail sites where the median page gets very few views (3 and 52).
- **Small or niche sites:** `my.brla.gov` (3.9M), `gisdata.brla.gov` (2.4M), `gis.brla.gov` (0.9M), `311.brla.gov` (0.7M), `checkbook.brla.gov` (181K), `budget.brla.gov` (126K) and `data.ebrgis.opendata.arcgis.com` (about 10.6K).
- **Skew within sites:** Page views per row are heavily skewed. For example, `www.brla.gov` has a median of 19 views per page against a maximum of about 17.6M. Averages are therefore dominated by a few pages, so use medians when comparing sites.
**2. Engagement behaviour (bounce and exit rate)**
These averages are unweighted across page rows.
- **High bounce and exit:** `311.brla.gov` (57% bounce, 56% exit), `gis.brla.gov` (28% / 42%), `city.brla.gov` (26% / 35%) and `brgov.com` (23% / 32%).
- **Mid:** `www.brla.gov` (18% / 25%), `data.brla.gov` (10% / 16%) and `my.brla.gov` (11% bounce, 37% exit).
- **Near-zero:** `gisdata.brla.gov` (0.3% / 0.4%), `checkbook.brla.gov` (1% / 1%) and `budget.brla.gov` (1% bounce, 10.5% exit). These look like tracking or measurement artifacts, or API-style and utility pages, so check them before reading them as "good engagement".
**3. Lifecycle (by YEAR coverage)**
- **Legacy or retired:** `brgov.com` runs 2014–2022, `gis.brla.gov` 2015–2021, `data.ebrgis.opendata.arcgis.com` 2016–2022, and `311.brla.gov` 2016–2023.
- **Current across the period:** `www.brla.gov` and `city.brla.gov` (2018–2024), `data.brla.gov`, `my.brla.gov` and `budget.brla.gov` (2015–2024).
- **New:** `gisdata.brla.gov` (2023–2024) and `checkbook.brla.gov` (2022–2024).
**Practical segmentation:** group first by WEBSITE, then by the three tiers above (dominant, long-tail content, niche services), and add YEAR for trends. Because of the skew, you could also bucket individual pages by page-view volume, such as top pages versus the long tail, within each site. PAGE PATH and PAGE URL are too granular to group on directly, since a site can have thousands of distinct pages.