Apropos of nothing: some numbers about millennial temples. To justify its place at JI, let’s call it an exercise in evaluating an agent’s perspective.
In 1856, Brigham Young famously proclaimed that
to accomplish this work there will have to be not only one temple but thousands of them, and thousands and tens of thousands of men and women will go into those temples and officiate for people who have lived as far back as the Lord shall reveal. 
I am intrigued by the idea of “thousands of [temples].” How might Young have meant it?
- Figuratively: “thousands” = “a very big number” not necessarily “more than 1,999”;
- Allusively: “the hearts of thousands and tens of thousands shall greatly rejoice in consequence of…the endowment” (D&C 110:9); 
- Hyperbolically: Young sometimes got carried away (not least in 1856);
- Prophetically (better: “seer-ly”): he knew the quantity by revelation/vision; 
- Analytically: he—the colonizer, the organizer—assessed the situation quantitatively.
Something else? Background knowledge of Young supports all as plausible, but #5 would require documentation. Ignoring that, and in conjunction with the idea of millennial temple ordinances for everyone, ever, does “thousands” seem “reasonable?”
The endowment is the rate-limiter.  Assuming, as Young did, that genealogy would not impede, and using present-day ordinances, I estimate that 39 to 230 temples, with 23,000 endowment-room seats, 160,000 workers, and an industrial pace (0.47 names/man-hour), would suffice. [5, 6, 7] Currently, 38,000 “equivalent workers” process 0.22 names/man-hour in 150 temples with 30,000 seats. [8, 9] All these numbers are, of course, conjectural and approximate. 
Besides his possible distaste for a breakneck rate, in Young’s 1856 there were more ordinances and they often took longer—but not the 9x to 40x longer 2,000 temples suggest.  Further, Young established many agricultural towns unlikely, by design, to grow much. For those communities to participate—using 19th-century transportation—would require temples placed by geography rather than population; perhaps Young’s agrarianism guided his thinking on temples. On the other hand, at Nauvoo about 5,200 individuals received the initiatory/endowment in 42 working days; it would take 3,000 such temples to match the hypothesized millennial rate.  Then again, Salt Lake’s floor plan provided about 5x as much space for comparable work, suggesting only 600 temples. 
So… with one sentence and anachronistic data I’ve conjured and confused a putative aspect of Young’s outlook. It’s fun, but it’s not history, and I forgot what my point was.
 1856 Jun 22, Journal of Discourses, 3:372. The statement has been cited many times, especially since the advent of small temples, e.g., Elaine S. Dalton, “We Did This for You,” Ensign, 2004 Nov, p. 89; Ronald A. Rasband, “Ronald A. Rasband: Temple worship is an important pattern for each of you to set” Church News, 2009 Feb 10 (Devotional at Brigham Young University, 2009 Feb 10); Merrill J. Bateman, “The Dawn of a New Millennium,” speech at Brigham Young University, 2000 Jan 11. BCC had a related discussion a few weeks ago.
 The thousands/ten-thousands device appears multiple times in scripture (with figurative connotations; e.g., Deut 32:30, Ps 144:13, Micah 6:7, Rev 5:11, Alma 3:26). The phrase “tens of thousands” does not appear in the KJV but appears eight times in the Book of Mormon and D&C.
 Or, in secular scholar language, had some sort of experience that he interpreted as a vision or revelation.
 Things change, of course, if the 21,000-seat Conference Center converts to an endowment room or endowments start broadcasting to chapels (à la the Nauvoo and Palmyra dedications).
 Assume: (1) continuous operation, except Sundays; (2) abundant and efficient workers; (3) no genealogic difficulties (prior to the cited quote, Young explained that post-mortals would deliver genealogical data); (4) no ordinance duplication; (5) no changes in the ordinances; (6) no changes in temple design or scale; (7) 1.5 hr endowment, 20 s baptism, 20 s confirmation, 4.5 min initiatory, 1.5 min sealing to spouse or parents (all including transition time); (8) worker logistics (transportation, lodging, etc.) not be a problem; (9) 100 billion people still needing complete work; (10) maintenance scales linearly with time and traffic; (11) temples used only for proxy work; (12) each woman sealed to one spouse; each child sealed to one set of parents.
Maintenance: assume large temples operate 15 hrs/day, 5 days/wk for 24 wks and then close for 2 weeks’ maintenance. Arbitrarily assuming 0.5 capacity, it gets cleaned every 900 “capacity-hours.” Full (1.0) capacity, 24 hrs/day, 6 days/wk reaches 900 capacity-hours in 6.25 weeks. Assuming maintenance drops to 1 wk, 1 wk/7.25 wks = 14% of the time. In 1985 (last year this data reported) there were 5/17 million possible endowments: 0.25 capacity (12,000 seats, 7 endowments/seat/day, 5 days/wk, 48 wks/yr).
 For 100 billion 1.5-hour endowments in 1,000 years, with allowances for Sundays (14%) and maintenance (14% of remaining), 15,000 endowments need to finish every hour, which requires 23,000 endowment seats or, at 75 seats/room (see below), 310 endowment rooms. To keep up, baptisms need 86 fonts, initiatory, 390 stalls. Sealing is complicated since one person can have multiple relationships, but since we’ve assumed the genealogy is done, each woman (50 billion) will be sealed twice: to her husband and then to her children (all at once), which takes care of everyone in 100 billion ordinances: 390 sealing rooms. The temple range assumes 600-seat temples as the lower bound and 100-seat as the upper.
 Assuming a 1/10 worker/patron ratio for endowments, 25,000 people could maintain the 15,000 endowment/hr pace. For the others: 9 people/font: 770; 9/confirmations: 770; 7/initiatory: 2,700; 19/sealing: 3,500; Total: 33,000. Running 24 hrs/day, 6 dys/wk, with 4 8-hour shifts/day (6 hrs on-deck) and a 5-day workweek requires 130,000 workers on-deck every day with 20% (27,000) off: total = 160,000 staff. A small temple (2 each endowment, sealing, initiatory rooms; 50 seats/endowment room) could run with 810 workers. The efficiency (man-hours/endowment) is: (on-deck staff/hr) x (hours temples open/yr) / (endowments/yr).
 Small temples generally have 2 endowment rooms, each seating 50-60; large generally have 3-6, each seating 100-120. At present, there are about 150 temples in operation or planning, with about 400 endowment rooms. About 200 come from 50 temples with 3+ endowment rooms (assume 100 seats/room); for the other 200 rooms/100 temples, assume 50 seats. Total: 30,000 endowment seats, averaging 75/room. Other current facilities: 150 fonts (assume 1/temple); 600 initiatory stalls (assume 2/1 stalls/endowment rooms in 50 large temples, 1/1 in 100 small); 530 sealing rooms. Wikipedia, “Comparison of temples….”
 Assume 10 million endowments/year (twice the 1985 rate with 2.5x the seats) and twenty or so other values of dubious reliability; I can give the numbers if you want. The “equivalent worker”: three workers covering a six-hour shift by each working two hours is “equivalent” to one working six.
 Remembering the mushiness of the numbers… to reach millennial rates requires: tripling the effective number of people attending, doubling the time they stay, and doubling how fast they work while there.
 In attempting to gauge Young’s interior life, Leonard J. Arrington observed that though Young was “a bold and bluff leader” he was also “responsive to the chords of celestial music as mediated through a liturgical experience of profound significance to him….” Brigham Young: American Moses (Urbana: University of Illinois Press, 1986), 200-201.
Baptism for health and administering to the sick are examples of temple ordinances no longer performed. Nauvoo initiatory/endowments varied from one to five hours; by the 1920s they could take six to even nine but were then streamlined to three. Thomas Alexander, Mormonism in Transition: A History of the Latter-day Saints, 1890-1930 (Chicago: University of Illinois Press, 1996/1986), p. 300; George D. Smith, ed. An Intimate Chronicle: The Journals of William Clayton (Salt Lake City: Signature, 1991), 1845 Dec 10 – 1846 Feb 07.
With 100-seat temples operating at maximum throughput (24-hr days, 1.0 capacity), the additional work and length would have occupy 8.6x as much time per endowment (stipulated as 1.5 hrs) to require 2,000 temples; for 400-seat temples: 35x. The 19th-century Utah temples had three or four endowment rooms; assume 100 seats/rm. With, say, 12-hr days and 0.25 capacity, however, the factors drop to 1.1x and 4.3x, which are plausible.
 History of the Church (HC; 7:541-580) records, by my tabulation, 5,161 initiatory-endowments in 42 days (average: 120/day) within a window of 60; they didn’t work on Sundays, among other breaks. (There’s some uncertainty on the numbers.) The greatest volume per day was 512 (Feb 06); 2nd and 3rd were 295 and 234. GA Smith’s journal claimed 600+ on Feb 07, disagreeing with the 126 in the Seventies Record (HC 7:580).
 Guessing wildly, I’d say 1.25 Salt Lake Temple floors (150’x80′), are for initiatories through sealings, providing an area of about 15,000 ft2 versus Nauvoo’s 2,900 ft2. Assuming that throughput scales linearly with floor space, 120/day becomes 600. HC 7:542; DM McAllister, A Description of the Great Temple, Salt Lake City, and a Statement Concerning the Purposes for Which It Has Been Built [Pamphlet], illustrated edition (Salt Lake City: Bureau of Information, 1922); No author listed, House of the Lord: Historical and Descriptive Sketch of the Salt Lake Temple [Pamphlet], (Salt Lake City: George Q. Cannon and Sons, 1893).
In case you’re curious: The original post has, according to MS Word, 409 words; the notes have 1,578.